By Topic

Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

Cover Image Copyright Year: 2004
Author(s): Pollack, J.; Bedau, M.; Husbands, P.; Watson, R.; Ikegami, T.
Publisher: MIT Press
Content Type : Books & eBooks
Topics: Computing & Processing (Hardware/Software)
  • Print

Abstract

Artificial Life is an interdisciplinary effort to investigate the fundamental properties of living systems through the simulation and synthesis of life-like processes. The young field brings a powerful set of tools to the study of how high-level behavior can arise in systems governed by simple rules of interaction. Some of the fundamental questions include:What are the principles of evolution, learning, and growth that can be understood well enough to simulate as an information process?Can robots be built faster and more cheaply by mimicking biology than by the product design process used for automobiles and airplanes?How can we unify theories from dynamical systems, game theory, evolution, computing, geophysics, and cognition?The field has contributed fundamentally to our understanding of life itself through computer models, and has led to novel solutions to complex real-world problems across high technology and human society. This elite biennial meeting has grown from a small workshop in Santa Fe to a major international conference. This ninth volume of the proceedings of the international A-life conference reflects the growing quality and impact of this interdisciplinary scientific community.

  •   Click to expandTable of Contents

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Front Matter

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): i - xiv
      Copyright Year: 2004

      MIT Press eBook Chapters

      This chapter contains sections titled: Half Title, Title, Copyright, Contents, Preface View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Chemoton: A Model for the Origin of Long RNA Templates.

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 1 - 8
      Copyright Year: 2004

      MIT Press eBook Chapters

      How could genomes have arisen? Two models based on Ganti's Chemoton are presented which demonstrate that under increasingly realistic assumptions, template replication is facilitated without the need of enzymes. It can do this because the template state is stoichiometrically coupled to the cell cycle. The first model demonstrates that under certain kinetic and environmental conditions there is an optimal template length, i.e. one which facilitates fastest replication of the Chemoton. This is in contradiction to previous findings by Csendes who claimed that longer templates allowed more rapid replication. In the second model, hydrogen bonding, phosphodiester bonding and template structure is modeled, so allowing dimer and oligomer formation, hydrolysis and elongation of templates. Here, monomer concentration oscillates throughout the cell cycle so that double strands form at low monomer concentrations and separate at high monomer concentrations. Therefore, this simulation provides evidence that a protocell with Chemoton organization is a plausible mechanism for the formation of long templates, a notorious problem for studies of the origin of life. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Connecting Transistors and Proteins

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 9 - 14
      Copyright Year: 2004

      MIT Press eBook Chapters

      We connect transistors and proteins in two ways. The first is by showrng that they have much in common as fundamental devices of electronics and life. The second is by describing how an evolvable wiring of electronic devices can parallel the wiring of proteins into genetic regulatory networks. We then transform this connection into a methodology for the study of the evolutionary properties of circuits. The approach is based on the use of analog electronic circuit simulators. We present an example of implementation with the first results obtained. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Designed and Evolved Blueprints For Physical Self-Replicating Machines

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 15 - 20
      Copyright Year: 2004

      MIT Press eBook Chapters

      Self-replication is a process critical to natural and artificial life, but has been investigated to date mostly in simulation and in abstract systems. The near absence of physical demonstrations of self-replication is due primarily to the lack of a physical substrate in which self-replication can be implemented. This paper proposes a substrate composed of simple modular units, in which both simple and complex machines can construct and be constructed by other machines in the same substrate. A number of designs, both hand crafted and evolved, are proposed. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Emergent Robustness and Self-Repair through Developmental Cellular Systems

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 21 - 26
      Copyright Year: 2004

      MIT Press eBook Chapters

      Fault-tolerance and, even more, self-repair remain elusive properties in computing systems. In contrast, natural systems are often cited as examples of flexible, self-repairable systems. Such capabilities rely on many different aspects, but our hypothesis is that (adaptive) growth and cellularity are at the heart of these properties lacking so dearly in human artifacts. In this paper, we propose a simple cellular developmental system to backup through experimental results this hypothesis. First, we show that it is possible to evolve such systems to do specific tasks. Second, and more importantly, that these systems exhibit emergent robustness and self-repair capabilities through their own nature rather than specific design or directed evolution. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      On Self-referential Shape Replication in Robust Aerospace Vehicles

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 27 - 32
      Copyright Year: 2004

      MIT Press eBook Chapters

      We describe a multi-cellular shape replication mechanism implemented in a sensing and communication network, motivated by robust self-monitoring and self-repairing aerospace vehicles. In particular, we propose a self-referential representation (a “genome”), enabling self-inspection and selfrepair; an algorithm solving the problem for connected and disconnected shapes; and a robust algorithm recovering from possible errors in the “genome”. The presented mechanism can replicate combinations of predefined shapes and arbitrary shapes that self-organise in response to occurring damage. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      An Evolving and Developing Cellular Electronic Circuit

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 33 - 38
      Copyright Year: 2004

      MIT Press eBook Chapters

      A novel multi-cellular electronic circuit capable of evolution and development is described here. The circuit is composed of identical cells whose shape and location in the system is arbitrary. Cells all contain the complete genetic description of the final system, as in living organisms. Through a mechanism of development, cells connect to each other using a fully distributed hardware routing mechanism and differentiate by expressing a corresponding part of the genetic code thereby taking a specific functionality and connectivity in the system. The configuration of the system is found by using artificial evolution and intrinsic evolution at the schematic level is possible. Applications include the approximation of boolean functions and the evolution of a controller capable of navigating a Khepera robot while avoiding obstacles. The circuit is suited for a custom chip called POEtic, which is a generic platform to implement bio-inspired applications. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      An Environment for Simulating Kinematic Self-Replicating Machines

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 39 - 44
      Copyright Year: 2004

      MIT Press eBook Chapters

      A simulation framework is described in which a collection of particles moving in continuous two-dimensional space can be put together to build machines. A self replicating machine has been designed in this environment. lt is proposed that an environment such as this may facilitate the fabrication of self-replicating manufacturing systems. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Towards an evolutionary-developmental approach for real-world substrates

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 45 - 50
      Copyright Year: 2004

      MIT Press eBook Chapters

      Extending “body-brain” evolution to the real-world presents a number of difficulties due to conflicting idealizations between evolutionary and constructional models. Toward addressing this gap, we develop a simple model system to analyze the effects of undoing these idealizations. Preliminary experiments with this system show that high variability developmental substrates can influence evolutionary dynamics by causing ambiguities in selection. Furthermore the substrate can enable the evolution of adaptive responses to nondeterministic developmental effects. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Stepwise evolution of molecular biological coding

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 51 - 56
      Copyright Year: 2004

      MIT Press eBook Chapters

      Two principles that embody necessary characteristics of self-sustaining physical systems are espoused. These principles are then used to analyze the dynamics of coding self-organization in the process of nucleic acid sequencedependent protein synthesis. An artificial system is constructed in which catalysis of codon to amino-acid assignments is embedded hierarchically in protein sequence space. An example is provided of a system that evolves through execution of a coarse-grained binary code to execution of a more refined quarternary code. General implications for the construction of ALife systems are considered. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Once More Unto the Breach: Co-evolving a robot and its simulator

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 57 - 62
      Copyright Year: 2004

      MIT Press eBook Chapters

      One of the major challenges facing evolutionary robotics is crossing the reality gap: How to transfer evolved controllers from simulated robots to real robots while maintaining the behavior observed in simulation. Most attempts to cross the reality gap have either applied massive amounts of noise to the simulation, or conducted most or all of the evolution onboard the physical robot, an approach that can be prohibitively costly or slow. In this paper we present a new co-evolutionary approach, which we call the estimation-exploration algorithm. The algorithm automatically adapts the robot simulator using behavior of the target robot, and adapts the behavior of the robot using the robot simulator. This approach has four benefits: the process of simulator and controller evolution is automatic; it requires a minimum of hardware trials on the target robot; it could be used in conjunction with other approaches to automated behavior transferal from simulation to reality; and the algorithm itself is generalizable to other problem domains. Using this approach we demonstrate a reduction of three orders of magnitude in the number of evaluations on a target robot (thousands compared to only five). View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Closing the loop: Evolving a model-free visually guided robot arm

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 63 - 68
      Copyright Year: 2004

      MIT Press eBook Chapters

      Dynamic neuro-controllers are incrementally evolved for reaching and tracking movements by a physically simulated robot arm. An active vision system capable of controlling gaze direction and focus replaces the need for internal models of the robot. It is shown that closing the feedback loop allows for robot control being robust to changes in environment, sensors and robot-morphology. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Performance Evaluation of Neural Architectures for Sequential Tasks

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 69 - 74
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper considers a sequential task where the agent has to alternatively visit two rewarding sites to obtain food and water after first visiting the nest. To achieve a better fitness, the agent must have a working memory to reach the target position, must ignore irrelevant sensory inputs, and at a higher level, it has to deal with the non-Markovian order of sequential task in which the preceding state alone does not determine the next action. We compare the performance of neural control architectures in different environment settings and analyze the neural mechanisms and environment features exploited by the agents to achieve their goal. Simulation and experimental results using the Cyber Rodent robot show that a specific architecture outperformed the general recurrent controller. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evo1GL: Life in a Pond.

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 75 - 80
      Copyright Year: 2004

      MIT Press eBook Chapters

      In this work we present the first version of Evolgl, an artificial environment for the development and study of 3D artificial lifeforms. In this first phase on the development of the project we have focused in setting up a virtual world governed by its own laws, whose state had direct influence upon the artificial beings that inhabit it. Starting from the definition of this virtual world, we have designed a basic type of creature (Evolworm), and the genetic coding of its main characteristics. Evolutionary techniques are then used to evolve the morphological features and behavioral aspects of Evolworms. They must learn to be unfolded inside the world, escape from their enemies, find couple, and obtain food. All of this in absence of an explicitly defined fitness function. In the future we are using this environment to study some classical techniques in the evolutionary computation field, like niche programming, and promotion of junk code (introns). GA-P techniques are used to code the external appearance of the individuals (the texture), to let evolution end up with individuals adapted to be invisible in some zones of the world. The artificial system of vision, and the implementation of the worms' behavioral mechanisms so that their actions are provoked exclusively by the sensory information are still under development. At this moment, we have obtained distinct forms of evolworms, as well as different bosses of behavior that we describe in this article. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolving Plastic Neural Controllers stabilized by Homeostatic Mechanisms for Adaptation to a Perturbation

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 81 - 87
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper introduces our ongoing work consisting of evolving bio-inspired plastic neural controllers for autonomous robots submitted to various internal and external perturbations: transmission breaking, slippage, leg loss, etc. We propose a classical neuronal model using adaptive synapses and extended with two bio-inspired homeostatic mechanisms. We perform a comparative study of the impact of the two homeostatic mechanisms on the evolvability of a neural network controlling a single-legged robot that slides on a rail and that is confronted to an external perturbation. The robot has to achieve a required speed goal given by an operator. Evolved neural controllers are tested on long-term simulations to statistically analyse their stability and adaptivity to the perturbation. Finally, we perform behavioral tests to verify our results on the robot controlled with a sinusoidal input while a perturbation occurs. Results show that homeostatic mechanisms increase evolvability, stability and adaptivity of those controllers. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Stigmergic Cooperative Multi-Robot Control Architecture

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 88 - 93
      Copyright Year: 2004

      MIT Press eBook Chapters

      In nature, there are numerous examples of complex architectures constructed by relatively simple insects, such as termites and wasps, which cooperatively assemble their nests. A prototype cooperative multi-robot control architecture which may be suitable for the eventual construction of large space structures has been developed which emulates this biological model. Actions of each of the autonomous robotic construction agents are only indirectly coordinated, thus mimicking the distributed construction processes of various social insects. The robotic construction agents perform their primary duties srigmergically, i.e., without direct inter-agent communication and without a preprogrammed global blueprint of the final design. Communication and coordination between individual agents occurs indirectly through the sensed modifications that each agent makes to the structure. The global stigmergic building algorithm prototyped during the initial research assumes that the robotic builders only perceive the current state of the structure under construction. Simulation studies have established that an idealized form of the proposed architecture was indeed capable of producing representative large space structures with autonomous robots. This paper will explore the construction simulations in order to illustrate the multi-robot control architecture. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Crawling Out of the Simulation: Evolving Real Robot Morphologies Using Cheap, Reusable Modules

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 94 - 99
      Copyright Year: 2004

      MIT Press eBook Chapters

      A current issue in evolutionary robotics involves the coevolution of robot controllers and body morphologies built from modular parts. As part of ongoing research, a model for the evolution of the morphologies and neural network controllers of robots is described. Several robots are evolved for locomotion in simulation built from modules representing cheap, preexisting parts and one is physically built that has comparable behaviour with its original simulated version. The behaviour in simulation of such example robots is described. A brief comparison is made between the behaviour of a simulated robot whose design and behaviour has been evolved and its physically instantiated counterpart. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Functional Freeform Fabrication for Physical Artificial Life

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 100 - 105
      Copyright Year: 2004

      MIT Press eBook Chapters

      Solid freeform fabrication (SFF) allows 3D-printing of arbitrarily shaped structures, directly from computer-aided design (CAD) data. SFF has traditionally focused on printing passive mechanical parts. Advances in this technology and developments in materials science make it feasible to begin the development of a single, compact, robotic SFF system – including a small set of materials - which can produce complete, active, functional electromechanical devices - mobile robots, for instance. We are advancing steadily toward this goal, and successes thus far have included the freeform fabrication of zinc-air batteries, conductive wiring, flexure joints, and combinations of these with thermoplastic structures. Several essential functionalities – actuation, sensing, and control electronics - still remain to be realized before complete electromechanical systems can be produced via SFF. Conducting polymers (CP) are a class of materials which can be used to produce all of these. Several SFFcompatible CP processing methods have been identified, and actuators produced via one of these have been demonstrated. When coupled in a closed-loop with an evolutionary design system, the ability to produce robots entirely via SFF becomes a bridge between the physical and the simulated, giving artificial evolution a complete physical substrate of enormous richness to explore with little or no human involvement. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Quadrupedal Locomotion: GasNets, CTRNNs and Hybrid CTRNN/PNNs Compared

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 106 - 112
      Copyright Year: 2004

      MIT Press eBook Chapters

      Evolutionary Robotics seeks to use evolutionary techniques to create both physical and physically simulated robots capable of exhibiting characteristics commonly associated with living organisms. Typically, biologically inspired artificial neural networks are evolved to act as sensorimotor control systems. These networks include; GasNets, Continuous Time Recurrent Neural Networks (CTRNNs) and Plastic Neural Networks (PNNs). This paper seeks to compare the performance of such networks in solving the problem of locomotion in a physically simulated quadruped. The results in this paper, taken together with those of other studies (summarized in this paper) help us to assess the relative strengths and weaknesses of the these three different approaches. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolving Simulated Mutually Perceptive Creatures for Combat

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 113 - 118
      Copyright Year: 2004

      MIT Press eBook Chapters

      A fundamental obstacle in evolutionary simulations is the necessity of designing more complex simulations to elicit more complex behaviors. We use a combat-based fitness measure to attempt to circumvent this problem. We have designed a simulation that simultaneously evolves the brains and bodies of creatures for one-on-one combat in a threedimensional environment with realistic physics. By giving the creatures a rich sensorium, we allow them to react sensibly to each others' actions. We discuss the effective, elegant and diverse simulated fighters that emerge, and examine whether qualitatively greater evolutionary complexity arises. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Information Trade-Offs and the Evolution of Sensory Layouts

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 119 - 124
      Copyright Year: 2004

      MIT Press eBook Chapters

      In nature, sensors evolve to capture relevant information needed for organisms of a particular species to survive and reproduce. In this paper we study how sensor layouts may evolve in different environments and under pressure of different informational constraints. To do this we evolve sensor layouts for different environments and constraints using a fitness measure with weighted terms for redundancy and novelty, using, respectively, mutual information and Crutchfield's information metric. The results show how different sensor layouts evolve depending on the structure and complexity of the environment but also how selective pressure for redundancy or novelty might affect the design. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolving Flying Creatures with Path Following Behaviors

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 125 - 132
      Copyright Year: 2004

      MIT Press eBook Chapters

      We present a system which evolves physically simulated 3D flying creatures and their maneuvers. The creature is modelled as a number of articulated cylinders connected by triangular patagia in between. A creature's wing structure and its low-level controllers for straight flight are generated by an evolutionary algorithm. Then a feed-forward neural network is attached to the low-level controllers, and the connection weights of the network for a given trajectory are found by a genetic algonthm. We show that a control system sufficiently effective to allow aerial creatures to follow a complicated path can be achieved by two-step evolution process. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Behavioural Categorisation: Behaviour makes up for bad vision

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 133 - 138
      Copyright Year: 2004

      MIT Press eBook Chapters

      The performance of a mobile robot with a vision system is assessed in an everyday object categorisation task. The ability of the robot to arrive at a specified object, behavioural categorisation, is compared to the moment to moment results from the computations of its vision system, here called perceptual classification. It is found that the mobile robot using the vision system is significantly more accurate at behavioural categorisation than the underlying performance of the visual system's perceptual categorisation. This result is discussed as supporting the hypothesis that embodied systems using real time algorithms find that ‘fast, cheap’ visual systems are sufficient for their needs. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Evolution of Control and Adaptation in a 3D Powered Passive Dynamic Walker

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 139 - 145
      Copyright Year: 2004

      MIT Press eBook Chapters

      Humans demonstrate speed, efficiency, and adaptability when traveling over rugged terrain. Bipethi robots modeled on biological designs could replace or assist people working in difficult environments. However, current research into humanoid robots has not produced practical machines. This paper explores the use of evolutionary robotics to evolve a simulation of a ten-degree of freedom bipedal robot. This machine demonstrates many of the properties of human locomotion. By using passive dynamics and compliant tendons it conserves energy while walking on a flat surface. Its speed and gait can be dynamically adjusted and it is capable of adapting to discrepancies in both its environment and its bodies' construction. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolving Imitating Agents and the Emergence of a Neural Mirror System

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 146 - 151
      Copyright Year: 2004

      MIT Press eBook Chapters

      Imitation is a highly complex cognitive process, employing vision, perception, representation, memory and motor control. The underlying mechanisms that give rise to imitative behavior have attracted a lot of attention in recent years and have been the subject of research in various disciplines, from neuroscience to animal behavior and human psychology. In particular, studies in monkeys and humans have discovered a neural mirror system that demonstrates an internal correlation between the representations of perceptual and motor functionalities. In contradistinction to previous engineering-based approaches, we focus on the evolutionary origins of imitation and present a novel framework for studying the emergence of imitative behavior. We successfully develop evolutionary adaptive autonomous agents that spontaneously demonstrate imitative learning, facilitating a comprehensive study of the emerging underlying neural mechanisms. Interestingly, some of these agents are found to embody a neural “mirror” device analogous to those identified in biological systems. Further analysis of these agents' networks reveals complex dynamics, combining innate perceptual-motor coupling with acquired context-action associations, to accomplish the required task. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Comparison of Population Learning and Cultural Learning in Artificial Life Societies

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 152 - 157
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper examines the effect of the addition of cultural learning to a population of agents. Experiments are undertaken using an artificial life simulator capable of simulating population learning (through genetic algorithms) and lifetime learning (through the use of neural networks). To simulate cultural learning, the exchange of information through nongenetic means, a group of highly fit agents is selected at each generation to function as teachers which are assigned a number of pupils to instruct. Cultural exchanges occur through a hidden layer of an agent's neural network known as the verbal layer. Through the use of backpropagation, a pupil agent imitates the teacher's behaviour and overall population fitness is increased. We show that the addition of cultural learning is of great benefit to the population and that in addition, cultural learning causes the population to converge on a fixed lexicon describing its environment. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Computational Framework to Simulate the Coevolution of Language and Social Structure

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 158 - 163
      Copyright Year: 2004

      MIT Press eBook Chapters

      In this paper, a multi-agent computational model is proposed to simulate the coevolution of social structure and compositional protolanguage from a holistic signaling system through iterative interactions within a heterogeneous population. We implement an indirect meaning transference based on both linguistic and nonlinguistic information in communications, together with a feedback without direct meaning check. The emergent social structure, triggered by two locally selective strategies, friendship and popularity, has small-world characteristics. The influence of these selective strategies on the emergent language and the emergent social structure are discussed. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Boom and Bust: Environmental Variability Favors the Emergence of Communication

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 164 - 169
      Copyright Year: 2004

      MIT Press eBook Chapters

      Environmental variability has been proposed as an important mechanism in behavioral psychology, in ecology and evolution, and in cultural anthropology. Here we demonstrate its importance in simulational studies as well. In earlier work we have shown the emergence of communication in a spatialized environment of wandering food sources and predators, using a variety of mechanisms for strategy change: imitation (Grim, Kokalis, Tafti & Kilb 2000), localized genetic algorithm (Grim, Kokalis, Tafti & Kilb 2001), and partial training of neural nets on the behavior of successful neighbors (Grim, St. Denis & Kokalis 2002). Here we focus on environmental variability, comparing results for all of these mechanisms in a range of different environments: (a) environments with constant resources, (b) environments with random resources around the same mean, and (c) sine-wave variable environments with cycles of ‘boom and bust’. Communication, it turns out, is strongly favored by environmental variability on the pattern of ‘boom and bust’. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Expressing and Understanding Desires in Language Games

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 170 - 175
      Copyright Year: 2004

      MIT Press eBook Chapters

      We speak because we want to get certain things accomplished. In this study we present the simulation of a multiagent language game which takes this into account. In contrast to previous language game studies, our agents use reinforcement learning to learn a function assigning a value to every state of the game. This value, that tells the agent how desirable the state is, is used along with a forward model to select actions. The agent can select verbal and non-verbal actions, depending on whether speaking or manipulating the world directly is more likely to bring about the change which the agent desires. On top of these capabilites, we used two rule-based agents to train a language learner. The learner trains a forward model of context-dependent utterance effects, which he then uses to express his desires and understand the desires of other players. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Evolution of Affect-Related Displays, Recognition and Related Strategies

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 176 - 181
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper presents an ecologically motivated, bottom-up approach to investigating the evolution of expression, perception and related behaviour of affective internal states that complements game-theoretic studies of the evolutionary success of animal display. Our results show that the perception of displays related to affect greatly influences both the types of display produced and also the survival prospects of agents. Relative to agents that do not perceive rival agent internal state, affect perceivers prosper if the initial environment in which they reside provides numerous opportunities for interaction with other agents and resources. Conversely, where the initial environment with sparse resources does not allow for regular interaction, ability to perceive affect is not as facilitatory to survival. Furthermore, the agents evolve particular display strategies distorting the expression of affect and greatly influencing the proportion of affect perceiving to nonaffect perceiving agents over evolutionary time. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Language, Altruism and Docility: How Cultural Learning Can Favour Language Evolution

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 182 - 187
      Copyright Year: 2004

      MIT Press eBook Chapters

      Human language serves a number of different functions, one of the most prominent being communicating about relevant features of the environment. From the point of view of the speaker, if the communicated information is advantageous for the hearer but not for the speaker, this is an altruistic use of language, and, as such, it requires an explanation of its evolution. Simon 1990 proposed an explanation of altruism in humans based on the genetically inherited ‘docility’ of our species. In this paper we present artificial life simulations that apply Simon's ideas to the problem of the emergence of the altruistic use of language described above. From the point of view of evolutionary theory, the present work represents the first attempt to test Simon's ‘docility’ theory of altruism with agent-based computer simulations. From the point of view of language evolution, our simulations give an original explanation of (the altruistic aspect of) human language based on one of its most peculiar characteristic, namely, the fact that it is culturally transmitted. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolution of Plastic Sensory-motor Coupling and Dynamic Categorization

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 188 - 193
      Copyright Year: 2004

      MIT Press eBook Chapters

      We study the dynamic categorization ability of an autonomous agent that distinguishes rectangular and triangular objects. The objects are distributed on a two-dimensional space and the agent is equipped with a recurrent neural network that controls its navigation dynamics. As the agent moves through the environment, it develops neural states which, while not symbolic representations of rectangles or triangles, allow it to distinguish these objects. As a result, it decides to avoid triangles and remain for longer periods of time at rectangles. A significant characteristic of the network is its plasticity, which enables the agent to switch from one navigation mode to another. Diversity of this switching behavior will be discussed. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Song Grammars as Complex Sexual Displays

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 194 - 199
      Copyright Year: 2004

      MIT Press eBook Chapters

      We study the complex evolution of song grammars of the Bengalese finch. Their mating songs have the remarkable feature that they are described by finite-state automata. (Honda and Okanoya, 1999) In addition, it has been experimentally confirmed that complex songs are preferred by females and that the Bengalese finch's song is more complex than that of its ancestors. (Okanoya, 2002) These facts suggest that complex grammar-like systems may have evolved as a result of sexual selection. In order to explore this hypothesis, we study the communication between male and female finches, modeling their co-evolution by asymmetric finite-state automata. By introducing a complexity measure for song grammars, we study the effect of females' preferences. We observe that a gradual transition from lower complexity to higher complexity grammars is associated with the changing of male birds' courting strategy. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Analogies between Genome and Language Evolution

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 200 - 207
      Copyright Year: 2004

      MIT Press eBook Chapters

      The paper develops an analogy between genomic evolution and language evolution, as it has been observed in the historical change of languages through time. The analogy suggests a reconceptualisation of evolution as a process that makes implicit meanings or functions explicit. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Effects of Learning on the Evolution of Saussurean Communication

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 208 - 213
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper presents a computational framework for studying the influence of learning on the evolution of communication. In our model, an evolving population of learning agents is engaged in pairwise comunicative interactions. Simulation results show the genetic assimilation of trasmission behaviors as a consequence of saussurean learning. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Minimum cost and the emergence of the Zipf-Mandelbrot law

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 214 - 219
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper illustrates how the Zipf-Mandelbrot law can emerge in language as a result of minimising the cost of categorising sensory images. The categorisation is based on the discrimination game in which sensory stimuli are categorised at different hierarchical layers of increasing density. The discrimination game is embedded in a variant of the language game model, called the selfish game, which in turn is embedded in the framework of iterated learning. The results indicate that a tendency to communicate in general tenns, which is less costly, can contribute to the emergence of the Zipf- Mandelbrot law. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Modeling Multicellular and Tumorous Existence with Genetic Cellular Automata

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 220 - 225
      Copyright Year: 2004

      MIT Press eBook Chapters

      We model a population of cells using cellular automata with genetically-based rules. As in actual multicellular systems, each cell's state is based on environment and genetics. With imperfect reproduction and the accumulation of random mutations, tumorous behavior naturally emerges in a stochastic manner. We validate our model by reproducing results used to confirm other published models. We also demonstrate that our model exhibits both homeostatic and tumorous behavior using metrics based on clinical biopsy diagnosic techniques. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Whatever Emerges should be Intrinsically Useful

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 226 - 321
      Copyright Year: 2004

      MIT Press eBook Chapters

      The practical work presented in this paper uses a GA to evolve a cellular automata (CA) implementation of a binary numbers adder. One very useful way to compress the enormous search space and eventually find an optimal CA consists in adopting a macro-coding of the states and the rule table. It is further discussed how this work illustrates and defends our favorite position in the currently vivid epistemological debate around the notion of “emergence”. This position is Crutchfield's “intrinsic emergence” one, in which to say that a macro-property is emergent requires that this “property” supplies some mechanical and non-human observer with additional functionality. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Shaping collective behavior: an exploratory design approach

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 232 - 237
      Copyright Year: 2004

      MIT Press eBook Chapters

      In order to fulfill the true promise of decentralized, selforganizing intelligence, a major design problem has to be overcome. Designing the individual-level rules of behavior and interaction that will produce a desired collective pattern in a group of human or non-human agents is difficult because the group's aggregate-level behavior may not be easy to predict or infer from the individuals' rules. While the forward mapping from micro-rules to macro-behavior in self-organizing systems can be reconstructed using computational modeling techniques such as agent-based modeling, the inverse problem of finding micro-rules that produce interesting macro-behavior poses significant challenges, all the more as what constitutes “interesting” macro-behavior may not be known ahead of time. An exploratory design method is described in this paper. It relies on interactive evolution. We show how it can be used to discover new, “interesting” patterns of collective behavior when one does not know in advance what the system is capable of doing, a generic situation in the design of collective intelligent systems. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Updating Schemes in Random Boolean Networks: Do They Really Matter?

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 238 - 243
      Copyright Year: 2004

      MIT Press eBook Chapters

      In this paper we try to end the debate concerning the suitability of different updating schemes in random Boolean networks (RBNs). We quantify for the first time loose attractors in asyncrhonous RBNs, which allows us to analyze the complexity reduction related to different updating schemes. We also report that all updating schemes yield very similar cntical stability values, meaning that the “edge of chaos” does not depend much on the updating scheme. After discussion, we conclude that synchonous RBNs are justifiable theoretical models of biological networks. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Reducing Prejudice: A Spatialized Game-Theoretic Model for the Contact Hypothesis

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 244 - 249
      Copyright Year: 2004

      MIT Press eBook Chapters

      There are many social psychological theories regarding the nature of prejudice, but only one major theory of prejudice reduction: under the right circumstances, prejudice between groups will be reduced with increased contact. On the one hand, the contact hypothesis has a range of empirical support and has been a major force in social change. On the other hand, there are practical and ethical obstacles to any large-scale controlled test of the hypothesis in which relevant variables can be manipulated. Here we construct a spatialized model that tests the core hypothesis in a large array of game-theoretic agents. Robust results show that prejudicial strategies flourish in a segregated environment but are eliminated in an integrated environment. We take this to offer a new kind of support for the contact hypothesis. The model also suggests a deeper gametheoretic explanation for some of the social phenomena at issue. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Quasi-Stable States in the Iterated-Prisoner's Dilemma

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 250 - 255
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper describes the states of a heterogenous population of agents playing the Iterated-Prisoner's Dilemma. The interactions of the agents are governed by five interaction processes which range from highly localized interactions to complete mixing; while the evolution of the agents is governed by five adaptive processes which range from local processes to global processes. For certain combinations of interaction processes, adaptive processes and control parameters, the populations alternate between periods of cooperation and defection while spending relatively little time in between. In addition, even at high rates of mutation, the population does not degenerate into random play. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolving Memory: Logical Tasks for Cellular Automata

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 256 - 261
      Copyright Year: 2004

      MIT Press eBook Chapters

      We present novel experiments in the evolution of Cellular Automata (CA) to solve nontrivial tasks. Using a genetic algorithm, we evolved CA rules that can solve non-trivial logical tasks related to the density task (or majority classification problem) commonly used in the literature. We present the particle catalogs of the new rules following the computational mechanics framework. We know from Crutchfield et al (2002) that particle computation in CA is a process of information processing and integration. Here, we discuss the type of memory that emerges from the evolving CA experiments for storing and manipulating information. In particular, we contrast this type of evolved memory with the type of memory we are familiar with in Computer Science, and also with the type of biological memory instantiated by DNA. A novel CA rule obtained from our own experiments is used to elucidate the type of memory that one-dimensional CA can attain. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Complex Genetic Evolution of Self-Replicating Loops

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 262 - 267
      Copyright Year: 2004

      MIT Press eBook Chapters

      lt is generally believed that self-replication models constructed on cellular automata have quite limited evolutionary dynamics in both diversity and adaptative behavior. Contrary to this view, we show that complex genetic diversification and adaptation processes may occur in self-replicating loop populations. Applying newly developed tools for detailed genetic identification and genealogy tracing to evoloop populations, we uncovered a genotypic permutation space that expands combinatorially with replicator size. Within this space populations demonstrate broad behavioral diversity and non-trivial genetic adaptation, maximizing colony density while enhancing sustainability against other species. We also found a set of non-mutable subsequences enabling genetic operations that alter fitness differentials and promote long-term evolutionary exploration. These results reveal the amazing potential of cellular automata to re-create complex genetic evolution of selfreplicators in a simple, deterministic framework. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Redrawing the Boundary between Organism and Environment

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 268 - 273
      Copyright Year: 2004

      MIT Press eBook Chapters

      In this position paper, I argue that a fruitful, and as yet largely unexplored, avenue for artificial life research lies in modelling organisms (specifically, phenotypes) and environment as a single dynamical system. From this perspective, the origin and evolution of life is the progressive control of the dynamical system at a local level by constraints which are represented on an organism's genome. Such an approach shifts the focus of artificial life models away from the design of individuals, towards the interaction of an individual with its dynamic environment. It also blurs the boundary between organism and environment; the most important modelling distinction is no longer between an organism's body and its external environment, but rather between the genome (which is treated as an essentially symbolic structure) and phenotype-plus-environment combined. An evolutionary cellular automata system, called EvoCA, is introduced as a tool to explore these ideas. To demonstrate how this approach differs from traditional studies, two example applications of EvoCA are presented. One concerns sensor and effector evolution; the other concerns the origin of life, and in particular the evolution of genome-regulated self-stabilising dynamics. Advantages of the new approach are swmnarised, and sorne potential criticisms are considered. The paper concludes with a discussion of some implications of this shift in perspective. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Imitation and Inequity in Avoiding the Tragedy of the Commons

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 274 - 279
      Copyright Year: 2004

      MIT Press eBook Chapters

      We present a tragedy of the commons model in which individuals have no a priori knowledge of the immediate consequences of their actions: Each agent chooses actions using a simple neural network, which it gradually modifies to more closely imitate those of its wealthier neighbors. For a small commons size, the model leads neither to the tragedy of complete resource exhaustion nor to complete cooperation, but instead to the emergence of polarized ‘economic classes’ of poor and altruistic agents living amongst rich and greedy ones. The tragedy does emerge with larger commons sizes; we found that adding a degree of enforced local sharing among neighbors staves off tragedy there, and once again the economic stratification emerges. Though simple, the model displays a surprising range of dynamic behaviors at multiple temporal and spatial scales, as two fundamentally conflicting ‘ideologies’ war for control of agent behavior. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Quantum Coreworld: Competition and Cooperation in an Artificial Ecology

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 280 - 285
      Copyright Year: 2004

      MIT Press eBook Chapters

      Evolving systems, in principle, can exploit any tool in their genetic repertoire. This work specifies a computational chemistry consistent with the rules of Quantum Mechanics and a distributed artificial ecology that permits intervention by interested participants. A preliminary demonstration of two programs that use Quantum operations to authenticate one another is described. Competitors cannot impersonate the cooperating Quantum users without themselves using Quantum mechanics. Some limitations of this example are discussed. Readers can also visit the Quantum Coreworid ecology on the Internet at http://science.fiction.org. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Self-reproduction by glider collisions: the beehive rule

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 286 - 291
      Copyright Year: 2004

      MIT Press eBook Chapters

      We present a 3-value cellular automaton which supports self-reproduction by glider collisions. The complex dynamics emerge spontaneously in both 2d and 3d according to the 6-neighbor, k-totalistic, “beehive” rule; the 2d dynamics on a hexagonal lattice is examined in detail. We show how analogous complex rules can be found, firstly by mutating a complex rule to produce a family of related complex rules, and secondly by classifying rule-space by input-entropy variance. A variety of complex rules opens up the possibility of seeking a common thread to distinguish those few rules from the rest: an underlying principle of self-organization? View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The value of death in evolution: A lesson from Daisyworld

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 292 - 296
      Copyright Year: 2004

      MIT Press eBook Chapters

      The first Daisyworld model showed how simple interactions and varying growth rates of daisies could lead to a type of global homeostasis. This idea has been extended to include variable mortality rates as a form of natural selection. This paper shows that this extension can increase the temperature regulation and persistence of Daisyworid systems, even with higher overall mortality rates. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Flexible Balance of Evolutionary Novelty and Memory in the Face of Environmental Catastrophes

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 297 - 302
      Copyright Year: 2004

      MIT Press eBook Chapters

      We study the effects of environmental catastrophes on the evolution of a population of sensory-motor agents with individually evolving mutation rates, and compare these effects in a variety of control systems. The evolution of mutation rates must balance (i) the need for evolutionary “novelty,” which pushes mutation rates up, and (ii) the need for evolutionary “memory,” which pushes mutation rates down. We observe that an environmental catastrophe initially shifts the balance toward evolutionary novelty and causes mutation rates to evolve upwards. Then, as the population adapts to the new environment, the balance shifts back toward evolutionary memory and the mutation rate falls. These observations support the hypothesis that second-order evolution of the mutation maintains a flexibly shifting balance between evolutionary novelty and memory. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Kin-Selection: The Rise and Fall of Kin-Cheaters

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 303 - 308
      Copyright Year: 2004

      MIT Press eBook Chapters

      We demonstrate the existence of altruism via kin selection in artificial life and explore its nuances. We do so in the Avida system through a setup that is based on the behavior of colicinogenic bacteria: Organisms can kill unrelated organisms in a given radius but must kill themselves to do so. Initially, we confirm results found in the bacterial world: Digital organisms do sacrifice themselves for their kin—an extreme example of altruism—and do so more often in structured environments, where kin are always nearby, than in well-mixed environments, where the location of kin is stochastically determined. Having shown that helping one's kin is advantageous, we turn our attention to investigating the efficacy and implications of the strategies of kincheaters, those who receive help from kin but do not return it. Contrary to the expectations of current theory, we find that kin-cheaters outcompete kin-altruists. Our results cause us to question the stability of strategies that involve altruism between kin. Knowing that kin-altruism persists in biological systems, however, we search for, and find, conditions that allow kin-based altruism to persist in evolving systems despite the presence of kin-cheaters. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Homeostasis and Rein Control: From Daisyworld to Active Perception

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 309 - 314
      Copyright Year: 2004

      MIT Press eBook Chapters

      Homeostasis refers to the ability of organisms to maintain vital properties, such as body temperature, within a zone of viability, or of comfort, and the Gaia Hypothesis proposes that the Earth with its biota acts as a homeostatic whole. The Daisyworld model was proposed as one possible mechanism for providing this homeostatic regulation. Here a new and much simplified version of this model is presented, demonstrating that the combination of any ‘Hat function’ with any feedback, positive or negative, can lead to homeostasis through ‘Rein Control’. This principle is so general that it can be extended to other domains such as active perception, here demonstrated in a simulated robot. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Measuring Biological Complexity in Digital Organisms

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 315 - 321
      Copyright Year: 2004

      MIT Press eBook Chapters

      We define biological complexity as the genetic information that an organism has about its environment. We have significantly improved methods to measure complexity based on Sharmon Information Theory and the principle of mutation-selection balance from population genetics. The previous method of Adami et al. was a population-based measure; it examined the information content of all genomes corresponding to the same phenotype. This population-based method had inherent limitations to its ability to approximate complexity: it requires a full population that must be at equilibrium, genomes must be fixed-length, and the environment must have only a single niche. Our new method overcomes these difficulties because it is genome-based rather than population-based. We approximate the total information in a genome as the sum of the information at each locus. The information content of a position is calculated by testing all of the possible mutations at that position and measuring the expected frequencies of potential genes in the mutation-selection equilibrium state. We discuss how this method reveals the way information is embedded in the organism during the evolutionary process. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Role of Nearly Neutral Mutations in the Evolution of Dynamical Neural Networks

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 322 - 327
      Copyright Year: 2004

      MIT Press eBook Chapters

      The evolution of continuous time recurrent neural networks is increasingly being employed to evolve nervous systems for autonomous agents. Nonetheless, the picture of populations engaged in hill-climbing rugged fitness landscapes poses a problem of becoming trapped on a local hilltop. Developments in evolutionary theory and molecular biology have pointed to the importance of selective neutrality. The neutral theory claims that the great majority of evolutionary changes are caused not by Darwinian selection but by random drift of selectively neutral or nearly neutral mutants. However, with a few exceptions neutrality has generally been ignored in artificial evolution. This paper addresses the distribution of fitness effects of new mutations when evolving dynamical systems and provides evidence of an improved evolutionary search process when incorporating nearly-neutral drift. This is one of the most fundamental problems in artificial evolution, because it lies at the heart of maintaining a constant-innovative property. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Sustained Evolution from Changing Interaction

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 328 - 333
      Copyright Year: 2004

      MIT Press eBook Chapters

      We develop and analyze an agent-based simulation model the purpose of which is to achieve sustained evolution under sympatric conditions. Evolution is understood in the form of speciation, i.e. the emergence of reproductively isolated and functionally distinct populations. In the model, reproduction is expressed as a function of phenotype to phenotype interaction. A population with a given set of phenotypic interactions tends to adaptive stasis, whereas if the interaction space changes, so that new dimensions are added in the course of the process, new species can emerge. We show that this behavior is stable and leads to a persistent production of new species. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      See How She Runs: Towards Visualising Artificial Red Queen Evolution

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 334 - 339
      Copyright Year: 2004

      MIT Press eBook Chapters

      Since Darwin originally proposed that one of the primary drives in evolution towards increased complexity might be biotic competition, the search for examples of this phenomenon has been a major objective for students of evolution. The Red Queen hypothesis proposes that this competition is never-ending; as an organism or species improves in some manner relative to its competitors, these competitors are also improving. This results in an evolutionary “arms race” in which neither side maintains its advantage for long. Examples of the Red Queen effect from the natural world can be found, but it is perhaps in artificial evolutionary systems that there is the greatest scope for gaining a deeper understanding of the nature of evolutionary change under such circumstances, whether gradualistic or punctuational for example. Here we present a model of coadaptation and introduce a visualisation method for tracking changes in the fitness landscape and the population's location on it. This enables the Red Queen's endless run within the model to be visualised. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Sexual reproduction and Muller's ratchet in digital organisms

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 340 - 345
      Copyright Year: 2004

      MIT Press eBook Chapters

      The evolution of sexual reproduction has long been a major problem in biology. According to one theory, sex opposes the fitness-destroying process of Muller's ratchet, which occurs by the stochastic loss of high-fitness genotypes in small populations. Sex opposes the ratchet by allowing genotypes with different deleterious mutations to produce mutation-free offspring. We used the Avida digitalevolution software to investigate sex in relation to Muller's ratchet. Populations of digital organisms mutated, competed, and evolved in a complex environment. Populations were either asexual or sexual; in the latter case, parental genomes recombined to produce offspring. We also varied genomic mutation rates and population sizes, which at extreme values often caused mutational meltdowns and population extinctions. Our results demonstrate that sex is advantageous for population survival under some conditions. However, differences in extinction probabilities were usually small, occurred over a narrow range of mutation rates and population sizes, and the advantage of sex for population survival required many generations. Also, the mean fitness of surviving asexual populations was often greater than in sexual populations. This last result indicates the need for work that compares the statistical distribution of mutational effects and epistatic interactions in asexual and sexual populations. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Chaotic Population Dynamics and the Evolution of Aging

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 346 - 351
      Copyright Year: 2004

      MIT Press eBook Chapters

      According to accepted evolutionary theories, aging has evolved as a side-effect of strong selection pressure for early fertility, despite the fact that it has no adaptive value of its own. I have argued elsewhere that recent experimental results make these theories untenable, and that there is now a broad array of evidence indicating that aging has evolved as an adaptation, selected for its own sake. To explain nature's preference for aging is a substantial theoretical challenge. The classical Weismann hypothesis, “making room for the young,” faits because the benefit to the population accrues in the form of enhancement to the rate of increase of population average fitness, while the cost affects individual fitness directly and efficiently. In multi-level selection models, the aging genes are lost before their benefit can accumulate. I propose here that aging has evolved based on a different benefit: its contribution to demographic homeostasis. I argue that population dynamics are inherently chaotic, and that the stable ecosystems that we commonly observe in nature are a highly evolved phenomenon. Natural selection for population homeostasis is far more efficient than selection for rate of evolution because chaotic population dynamics can be lethal on a time scale of just a few generations, while enhanced rate of evolution takes far longer to affect population mean fitness. My thesis is that aging can evolve based on its ability to damp population fluctuations. For illustration, I offer an individual-based computational model that reproduces chaotic population dynamics with a delayed-feedback logistic equation. Genes for aging emerge handily. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Role of Non-Genetic Change in the Heritability, Variation, and Response to Selection of Artificially Selected Ecosystems

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 352 - 357
      Copyright Year: 2004

      MIT Press eBook Chapters

      A response to selection on the level of the ecosystem has been demonstrated in artificial selection experiments, and poses interesting challenges to concepts of heritability, variation and phenotype in biological systems. We use ecosystems modeled as Lotka-Volterra competition systems, and subject to an ecosystem-level selection process, to illustrate and discuss the potential, and possible mechanisms, for ecosystem-level evolution without genetic change of the component species. A limited positive response to selection is demonstrated by the selection of alternative stable ecosystem states. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Ecolab, Webworld and self-organisation

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 358 - 363
      Copyright Year: 2004

      MIT Press eBook Chapters

      Ecolab and Webworld are both models of evolution produced by adding evolution to ecological equations. They differ primarily in the form of the ecological equations. Both models are self-organised to a state where extinctions balance speciations. However, Ecolab shows evidence of this self-organised state being critical, whereas Webworld does not. This paper examines the self-organised states of these two models and suggest the likely cause of the difference. Also the lifetime distribution for a mean field version of Ecolab is computed, showing that the fat tail of the distribution is due to coevolutionary adaption of the species. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Tierra's missing neutrality: case solved.

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 364 - 368
      Copyright Year: 2004

      MIT Press eBook Chapters

      The concept of neutral evolutionary networks being a significant factor in evolutionary dynamics was first proposed by Huynen et al. about 7 years ago. In one sense, the principle is easy to state — because most mutations to an organism are deleterious, one would expect that neutral mutations that don't affect the phenotype will have disproportionately greater representation amongst successor organisms than one would expect if each mutation was equally likely. So it was with great surprise that I noted neutral mutations being very rare in a visualisation of phylogenetic trees generated in lien-a, since I already knew that there was a significant amount of neutrality in the Tierra genotype-phenotype map. It turns out that competition for resources between host and parasite inhibits neutral evolution. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Drastic Changes in Roles of Learning in the Course of Evolution

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 369 - 374
      Copyright Year: 2004

      MIT Press eBook Chapters

      An interaction between evolution and learning called the Baldwin effect is known as the two-step evolutionary scenario caused by the balances between benefit and cost of learning in general. However, little is still known about dynamic evolutions on these balances in complex environments. Our purpose is to give a new insight into the benefit and cost of learning by focusing on the quantitative evolution of phenotypic plasticity under the assumption of epistatic interactions. For this purpose, we have constructed an evolutionary model of quantitative traits by using an extended version of Kauffman's NK fitness landscape. Phenotypic plasticity is introduced into our model, in which whether each phenotype is plastic or not is genetically defined and plastic phenotypes can be adjusted by learning. The simulation results have clearly shown that the drastic changes in roles of learning cause the three-step evolution of the Baldwin effect (Suzuki and Arita, 2003) and also cause the evolution of the genetic robustness against mutations. We also conceptualize four different roles of learning by using a hill-climbing image of a population on a fitness landscape. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Niche Construction and the Evolution of Complexity

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 375 - 380
      Copyright Year: 2004

      MIT Press eBook Chapters

      An individual-based model of the process of niche construction is presented, whereby organisms disturb the environment experienced by their neighbours. This disturbance in local conditions creates a niche that potentially could be filled by another species (which would then create still more niches and so on). The model is unique in allowing the complexity of the organisms—measured by the number of genes they possess in order to be well adapted to their local environment—to evolve over time, and is therefore the first model with which it is possible to study the contribution of niche construction to the evolution of organism complexity. Results of experiments demonstrate that the process of niche construction does indeed introduce an active drive for organisms with more genes. This is the first explicit example of a model which possesses an intrinsic drive for the evolution of complexity. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Model for Exploring Genetic Control of Artificial Amoebae

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 381 - 386
      Copyright Year: 2004

      MIT Press eBook Chapters

      We develop a computer simulation of several cellular processes seen in amoebae, including the production and regulation of proteins via a genome; the production, release, and destruction of diffusible chemicals; and regulated chemotaxis through a lattice environment facilitated by the interactions of proteins and diffusible chemicals. We also test this model by adapting biological situations to this model to evaluate its ability to model genetic networks and genetically regulated chemotaxis. The model will be used to simulate evolution in artificial amoebae to produce behavior seen in biological organisms such as Dictyostelium discoideum. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Asymmetric cell division and its integration with other developmental processes for artificial evolutionary systems

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 387 - 392
      Copyright Year: 2004

      MIT Press eBook Chapters

      Artificial evolutionary techniques are more and more coupled with mechanisms abstracted from developmental biology. Artificial cells endowed with genetic regulatory networks were used to evolve and develop simulated creatures. This paper reports on the evolution of a simple moving creature using developmental mechanisms such as asymmetric cell division, genetic regulation and cell adhesion and physical interactions between cells. Surprisingly, artificial creatures were evolved able to move using only genetic regulatory networks without the need to employ neural controllers. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Functional Model of Cell Genome

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 393 - 398
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper is concerned with a model of the cell genome called Artificial Genome, that tries to model some aspects of the cell cycle, in particular those related to gene expression, cell differentiation and cell growth. The functioning of the model during interphase and mitosis is explained in detail through an example, that shows how the four functional categories of the Artificial Genome (Functions, Code, Data and Buffer) interact to determine the phenotype. The capacity of the model of generating phenotypical patterns, represented as 2-dimensional shapes, is explored through a simulation, that evolves in 9 cycles a cell to become a small face made up of 132 cells. Finally some parlallels between the Artificial Genome and the natural one are discussed. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Asynchronous Dynamics of an Artificial Genetic Regulatory Network

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 399 - 403
      Copyright Year: 2004

      MIT Press eBook Chapters

      The synchrony / asynchrony dichotomy prevalent in models of genetic regulatory networks can be replaced by a parameter, s, which is the probability of a node being updated in a single time step. Here we apply the idea of such parameterized synchrony to study the dynamics of the genetic regulatory network extracted from an artificial genome model. We find that the relationship between degree of synchrony and the number of limit cycles is not linear. The number and length of limit cycles peaks at intermediate values of s. The proportion of state space explored and the length of transient trajectories also follows this pattern. The richer behavior found at intermediate values of the synchrony parameter is much more characteristic of biological systems than either full synchrony or complete asynchrony. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Small World and Scale-Free Network Topologies in an Artificial Regulatory Network Model

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 404 - 409
      Copyright Year: 2004

      MIT Press eBook Chapters

      Small world and scale-free network topologies commonly exist in natural and artificial systems. Many mechanisms for producing these topologies have been presented in the literature. We present an artificial regulatory network model generated by a duplication / divergence process on a randomly generated genetic string and show that networks with small world and scale-free topologies can be produced with some regularity. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Inertia of Chemotactic Motion as an Emergent Property in a Model of an Eukaryotic Cell

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 410 - 414
      Copyright Year: 2004

      MIT Press eBook Chapters

      Chemotaxis is widely seen in many biological systems. Among them amoebic cells from unicellular slime molds to immune cells are believed to directly sense chemical gradients. Here, we construct a model of amoebic cell by taking account of the chemical kinetics as well as a cellular body. The model is composed of discrete grids and a set of rules which define chemical and motional events on each grid. The model can explain the observed features of the cellular locomotion. We find that the simulated cell tends to keep the direction of motion, which reminds us of “inertia” of motion in Newtonian dynamics. The averaged motion of amoebic cells approximately obeys an “underdamped” equation of motion for a short time scale. “Inertia” of chemotactic motion is an emergent property of the system where motion and the signal processing are strongly coupled to each other. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Phenotypic Variability in Canalized Developmental Systems

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 415 - 420
      Copyright Year: 2004

      MIT Press eBook Chapters

      The evolution of nervous systems has been relatively conservative despite a large diversity of behavior. Novel behaviors appear to be produced during evolution by changing synaptic connectivity (Edwards and Palka 1991, Shaw and Meinertzhagen 1986, and Nishikawa et al. 1992). Due to a process of canalization, in which the development of organisms becomes buffered against genetic and environmental changes, phenotypic variation is reduced (Waddington 1942). Therefore, we explore phenotypic variability in simulations of neural development in which the developmental pathways have become canalized. Variation is determined under point mutations and gene knockouts. Although a genetic regulatory network utilizing activation and controlling a single process of neural development is evolvable and can become canalized, large genetic perturbations do not result in increased phenotypic variability as expected from related work (Bergman and Siegal 2003). View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Self-repairing and Mobility of a Simple Cell Model

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 421 - 426
      Copyright Year: 2004

      MIT Press eBook Chapters

      The evolution of a mobile cell system is studied here. While a primary definition of life is self-reproduction, mobility is also a major characteristic. Typically, most alife models assume cell mobility a priori. Here, using a development of a cell model due to Varela and McMullin, we show the emergence of cell mobility. A balance between cell repair and mobility is demonstrated. The introduction of a new functional particle is a critical part of the present model. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evaluating an Evolutionary Approach for Reconstructing Gene Regulatory Networks

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 427 - 432
      Copyright Year: 2004

      MIT Press eBook Chapters

      Reconstructing networks from (partial) incomplete data is a general problem in biology. We use an evolutionary approach in an artificial network creation and reconstruction framework to investigate limitations of gene expression network inference from simulated microarray data. For this, the simulated dynamics of the evolved networks are optimized to fit the target dynamics. Evolving networks with similar dynamics is not as difficult as comparing the resulting network topologies to the original network to be reconstructed. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Bonding as an Emergent Phenomenon in an Abstract Artificial Chemistry

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 433 - 437
      Copyright Year: 2004

      MIT Press eBook Chapters

      In this article we describe an implementation of a very simple toy artificial chemistry originally proposed as a “Gedanken” model by Groß and McMullin. They predicted that this model would have a number of interesting properties including emergent bonding, and top-down constraints on bondstability. We found that not all of the original claims could actually be verified. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Evolution of Robust Developmental Neural Networks

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 438 - 443
      Copyright Year: 2004

      MIT Press eBook Chapters

      We present the first evolved solutions to a computational task within the Neuronal Organism Evolution model (Norgev) of artificial neural network development. These networks display a remarkable robustness to external noise sources, and can regrow to functionality when severely damaged. In this framework, we evolved a doubling of network functionality (double-NAND circuit). The network structure of these evolved solutions does not follow the logic of human coding, and instead more resembles the decentralized dendritic connection pattern of more biological networks such as the C. ele gans brain. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Functional Self-Reproducing Cell in a Two-Dimensional Artificial Chemistry

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 444 - 449
      Copyright Year: 2004

      MIT Press eBook Chapters

      We show how it is possible to make a self-reproducing cell in an artificial chemistry by surrounding a replicating molecule with a semi-permeable membrane. The molecule can carry an arbitrary amount of information, encoded in a material form as a sequence of bases, as in DNA. The cells produce enzymes through a decoding of their base sequence, and these enzymes trigger reactions essential to the cell's survival. Earlier work in a similar artificial chemistry showed that replicators free in solution could obtain no survival advantage from producing enzymes; here we show that when surrounded by a membrane the replicators can obtain an advantage. We show that the cells reliably reproduce over many generations under environmental pressure for resources. By creating cells in a material-based artificial chemistry we hope that the system might have the potential for open-ended, creative evolution. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Metabolic closure in (M,R) systems

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 450 - 455
      Copyright Year: 2004

      MIT Press eBook Chapters

      The work of Robert Rosen, related to metabolic networks called (MR) systems is reviewed and clarified. We study the algebraic formulation of (M,R) systems particularly the mapping β, which encapsulates Rosen's solution to the problem of metabolic closure and infinite regress. We construct an arithmetical example of an (MR) system and also an (MR) system based on a three-step minimal metabolism. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Flows of information in spatially extended chemical dynamics

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 456 - 460
      Copyright Year: 2004

      MIT Press eBook Chapters

      A continuity equation for information is presented, involving flows in both space and scale in chemical pattern formation systems. The flows are connected to thermodynamic properties of the system. Information leaves the system (or is destroyed) at the smallest length scales, which corresponds to entropy production. Information enters the system at the largest length scales when the system is open to an inflow of Gibb's free energy. The continuity equation describes how information can be aggregated at different scales and positions during the pattern formation process. The formalism is applied to the Gray-Scott model, exhibiting self-replicating spots in a spatially extended system. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Cellular Dynamics in a 3D Molecular Dynamics System with Chemistry

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 461 - 465
      Copyright Year: 2004

      MIT Press eBook Chapters

      We present a three-dimensional model of the formation of simple protocellular structures. The model is based on an earlier lattice artificial chemistiy due to Ono and Ikegami which consisted of a primitive metabolic system built on an artificial chemistry. This model computed the interactions of simple amphiphilic molecules which organized into membranelike structures. The current model, however, treats space as continuous rather than a lattice. Moreover, although forces between atoms are computed in a more realistic manner, an adaptive method of computing intermolecular forces allows for efficient computation. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Lipidia: An Artificial Chemistry of Self-Replicating Assemblies of Lipid-like Molecules

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 466 - 471
      Copyright Year: 2004

      MIT Press eBook Chapters

      Lipidia is a new simulation system that is related to the “Lipid World” scenario for the origin of life. Lipidia allows for conducting experiments with a population of assemblies containing lipid-like molecules on a two dimensional grid. The dynamics of the assemblies is modelled using the Graded Autocatalysis Replication Domain (GARD) model. New experiments using a finite environment model with GARD were conducted with Lipidia. The experiments show that more self-replicating assembly species appear when using a model of finite environment than when using a model of infinite environment. In many species the number of individuals increases as well. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Towards the Simulation of Reaction Networks in Astrochemistry

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 472 - 477
      Copyright Year: 2004

      MIT Press eBook Chapters

      The aim of this paper is to apply artificial chemistry (AC) to the modelling of astrochemistry. More generally, we have attempted to construct a computational model of ‘a’ minimal real chemistry in an object oriented (OO) approach (UML, Java). To this end we map astrochemistry into the ACs theoretical framework thanks to the developed OO model. The OO approach has permitted us to separate the actual and logical representation of chemical structure. To perform this mapping we identify and simplify features from astrochemistry to tackle the issues of, representation and canonization, reaction rules, and meta-dynamics for a suitable level of abstraction. In the search for a chemical criterium for existence relations between objects we have used chemical thermodynamics. The result is an abstract, simplified model of a red chemical dynamical reaction network which conceptually captures real chemistry in general. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Homochirality as Fixed Point of Prebiotic Chemistry

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 478 - 483
      Copyright Year: 2004

      MIT Press eBook Chapters

      The purpose of this work is to highlight the ability of a pre-biotic system of activation/polymerization/depolymerization ( APD) of amino acids to have non-racemic fixed points, similarly to Frank's model, and thus to evolve spontaneously toward homochirality. Chemical kinetic simulations of reaction sets, from simple isolated polymerization systems to complete APD systems in presence of inversion reactions, are developed in order to understand the mechanisms of amplification of asymmetry. The results emphasize the emergence of autocatalysis thanks to the synergetic action between epimer-ization and APD reactions, allowing spontaneous symmetry breaking from racemic state. The APD system appears to be an original nonequilibrium chemical system model, as an extension of the Frank's model based on prebiotically relevant chemical reactions. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      An Evolvable Artificial Chemistry Featuring Continuous Physics and Discrete Reactions

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 484 - 488
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper describes an artificial chemistry featuring atoms and molecules moving and colliding in a continuous manner in a viscous fluid filling a 2D cellular space. Chemical reactions are mappings of discrete cellular configurations to parameterized actions on atoms. Actions allow atom creation and destruction, bonding and unbonding to make and break molecules, orientation, type change, and propulsion. Actions are easily added in this extensible model. An example involving a complex “foraging” reaction is provided as a demonstration of the capabilities of the framework. The reaction rules can be evolved by a genetic algorithm to exhibit a desired set of reactions. A portion of the foraging reaction was evolved to demonstrate this. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      The Role of RNA Editing in Dynamic Environments

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 489 - 494
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper presents a computational methodology based on Genetic Algorithms with Genotype Editing (GAE) for investigating the role of RNA editing in dynamic environments. This model is based on genotype editing characteristics that are gleaned from RNA editing processes as observed in several organisms. We have previously expanded the traditional Genetic Algorithm (GA) with artificial editing mechanisms (Rocha, 1995, 1997), and studied the benefits of including straightforward Genotype Editing in GA for several machine learning problems (Huang and Rocha, 2003, 2004). Here we show that genotype editing also provides a means for artificial agents with genotype/phenotype mappings descriptions to gain greater phenotypic plasticity. We simulate agents endowed with the ability to alter the edition of their genotype according to environmental context. This ability grants agents an adaptive advantage as genotype expression can become contextually regulated. The study of this genotype edition model in changing environments has shed some light into the evolutionary implications of RNA editing. We expect that our methodology will both facilitate determining the evolutionary role of RNA editing in biology, and advance the current state of research in Evolutionary Computation and Artificial Life. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Tangled Hierarchy of Graph-Constructing Graphs

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 495 - 500
      Copyright Year: 2004

      MIT Press eBook Chapters

      The traditional construction paradigm of machine and tape is reformulated in a functionally homogeneous space of directed graph structures. Hierarchy-based roles, normally appointed to actors in a construction process, are dissolved and replaced by symmetric, level-less engagement. The separation between static (information carrying) and active (information processing) structures, imposed by mandate of the rules or physics in earlier theoretical models, results instead purely firom graph topology. While encompassing traditional machine-tape paradigms as a special case, the formalism is shown to incorporate a wider class of construction relations. Exploiting its flexibility, a representation of a Turing machine is demonstrated, establishing computation universality. The concept of a “Tangled Construction Hierarchy” is introduced. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Graded Artificial Chemistry in Restricted Boundaries

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 501 - 506
      Copyright Year: 2004

      MIT Press eBook Chapters

      The question of the origin of life is addressed by artificial life research, particularly in the realm of artificial chemistry. Such artificial chemistry is described by our Graded Autocatalysis Replication Domain (GARD) model. GARD depicts an unorthodox scenario suggested for emergence of life — the ‘lipid world’. The model concerns molecular assemblies with mutual catalysis in an environment containing a plethora of molecular species. Many aspects of GARD were amply discussed. Here we concentrate on the importance of size constraints as depicted by the basic model and several of its variants. Occasional fission of a GARD assembly, which restricts the assembly size, is crucial for generating compositional quasi-stationary states (‘composomes’). In a spatial version of GARD, bounded environments yield spontaneous emergence of different ecologies. Limiting the size of a population of GARD assemblies gives rise to a complex population dynamics. The last example, with possible wider impact to chemistry and nano-technology, suggests that size limit can give rise to spontaneous symmetry breaking. This latter result is compared to the classic Frank's model for homo-chirality, which requires explicit inhibition. We conclude that size restrictions are fundamental in the field of origin of life and artificial life, not only in order to facilitate evolutionary processes, as previously suggested, but also, for augmenting the dynamics portrayed by different scenarios and models. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Spacial Representation for Artificial Chemistry Based on Small-World Networks

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 507 - 513
      Copyright Year: 2004

      MIT Press eBook Chapters

      A method to simulate molecular reactions with a graph whose nodes represent molecules or atomic modules and whose edges represent the bond/contact relation is proposed. The graph is updated by two actions: passive rewiring of contact edges that gives the graph a small-world property and active rewiring by active nodes' programs that rewire bond and contact edges. As examples, the replication of a chain molecule and the partitioning of a network are successfully simulated. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Behavioral Adaptive Autonomy. A milestone on the Alife route to AI?

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 514 - 521
      Copyright Year: 2004

      MIT Press eBook Chapters

      While central to robotics, biology and cognitive science, the concept of autonomy remains still difficult to make operative in the realm of Alife simulation models of cognitive agents. Its deep significance as a transition concept between life and cognition (a milestone on the Alife route to AI) remains obscured in the intricate relation between metabolic/constructive processes and behavioral adaptive processes in living systems. Within a naturalized and biologically inspired dynamical approach to cognition a definition of behavioral adaptive autonomy is provided: homeo-static maintenance of essential variables under viability constraints through self-modulating behavioral coupling with the environment, hierarchically decoupled from metabolic (constructive) processes. This definition allows for a naturalized notion of behavioral adaptive functionality (that defines a proper level of modelling within Alife), structurally and interactively emergent: the mapping of the agent-environment system's state space trajectories into the viability subspace of the essential variables of the organism. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Physiological Approach to the Generation of Artificial Life Forms

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 522 - 527
      Copyright Year: 2004

      MIT Press eBook Chapters

      Artificial Life has been an inspiration for Virtual Reality Art, which has proposed the design of imaginary life forms and has also used Artificial Life techniques in various installations. The creation of imaginary life forms has been so far limited to the external appearance of these creatures. We introduce a novel approach to the design of artificial life forms, which enables the description of their internal physiology from first principles, by using a simulation method known as qualitative physiology, derived from a well described Artificial Intelligence technique. We illustrate this framework by revisiting early work in Artificial Life and providing these virtual life forms with a corresponding physiology, so as to obtain complete living organism in virtual worlds. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Mechanistic and ecological explanations in agent-based models of cognition

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 528 - 533
      Copyright Year: 2004

      MIT Press eBook Chapters

      We argue that two styles of explanation—mechanistic and ecological—are needed in accounting for the behaviour of synthetic agents. An emphasis on mechanistic explanation in some current ALife models is identified, and parallels are drawn with issues in the philosophy of mind literature. We conclude that ecological or agent-level explanation does not come with representational baggage, and that mechanistic explanations cannot stand alone. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Empiricism in Artificial Life

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 534 - 539
      Copyright Year: 2004

      MIT Press eBook Chapters

      Strong artificial life research is often thought to rely on Alife systems as sources of novel empirical data. It is hoped that by augmenting our observations of natural life, this novel data can help settle empirical questions, and thereby separate fundamental properties of living systems from those aspects that are merely contingent on the idiosyncrasies of terrestrial evolution. Some authors have questioned whether this approach can be pursued soundly in the absence of a prior, agreed-upon definition of life. Here we compare Alife's position to that of more orthodox empirical tools that nevertheless suffer from strong theory-dependence. Drawing on these examples, we consider what kind of justification might be needed to underwrite artificial life as empirical enquiry. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Using the Universal Similarity Metric to Model Artificial Creativity and Predict Human Listeners Response to Evolutionary Music

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 540 - 545
      Copyright Year: 2004

      MIT Press eBook Chapters

      In this paper we present a new technique for modeling Artificial Creativity in Evolutionary Music (EM) systems and predicting how appealing musical pieces are to human listeners. We use a k-Nearest Neighbor classifier where we approximate the Information Distance between the new, unclassified, musical piece and a corpus of observed musical pieces rated by the user with the Universal Similarity Metric. We approximate the Information Distance with two different methods, using standard binary compression of MIDI files, and using MP3 encoding of raw audio streams. Our experiments indicate that the universal similarity metric can be used to discriminate between music that do and do not appeal to human listeners. Even though classification results is not perfect, it performs significantly better than the random baseline and when we combine the predictions made independently by the MIDI and MP3 classifiers, we obtain an even higher classification accuracy, ranging up to 77% on the test set. These results is in the same range as our results in predicting the aesthetic value of visual art, which indicates that the Universal Similarity Metric is a very general and versatile approach to modeling Artificial Creativity. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Playing Music by Conducting BOID Agents — a Style of Interaction in the Life with A-Life

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 546 - 550
      Copyright Year: 2004

      MIT Press eBook Chapters

      This paper presents an interactive installation that employs flocking algorithms to produce music and visuals. The user's motions are captured by a video camera and influence the flocks behaviour. Each agent moving in a virtual 3D space controls a MIDI instrument whose playing style depends on the agent's state. In this system, the user acts as a conductor influencing the flock's musical activity. In addition to gestural interaction, the acoustic properties of the system can be modified on the fly by using an intuitive GUI. The acoustical and visual output of the system results from the combination of the flock's and user's behaviour. It therefore creates on the behavioural level a mixing of natural and artificial reality. The system has been designed to run an a variety of different computational configuration ranging from small laptops to exhibition scale installations. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      An Evolutionary Approach to Complex System Regulation Using Grammatical Evolution

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 551 - 556
      Copyright Year: 2004

      MIT Press eBook Chapters

      Motivated by difficulties in engineering adaptive distributed systems, we consider a method to evolve cooperation in swarms to model dynamical systems. We present an information processing swarm model that we find to be useful in studying control methods for adaptive distributed systems. We attempt to evolve systems that form consistent patterns through the interaction of constituent agents or particles. This model considers artificial ants as walking sensors in an information-rich environment. Grammatical Evolution is combined with this swarming model as we evolve an ant's response to information. The fitness of the swarm depends on information processing by individual ants, which should lead to appropriate macroscopic spatial and/or temporal patterns. We discuss three primary issues, which are tractability, representation and fitness evaluation of dynamical systems and show how Grammatical Evolution supports a promising approach to addressing these concerns. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Analyzing Evolved Fault-Tolerant Neurocontrollers

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 557 - 562
      Copyright Year: 2004

      MIT Press eBook Chapters

      Evolutionary autonomous agents whose behavior is determined by a neurocontroller “brain” are a promising model for studying neural processing. Nevertheless, they are missing an important quality prevalently found in all levels of natural systems, fault-tolerance, the lack of which results in overly simplistic neurocontrollers. We present a way of modifying a given evolutionary process for encouraging the creation of neurocontrollers that manifest high levels of fault-tolerance, using both direct and incremental evolutions. The evolved neurocontrollers are more robust not only against the faults introduced during the evolutionary process, but also against much more extreme ones. This robustness poses a great challenge for an analysis of the workings of the neurocontrollers, the latter being the focus of this paper: We utilize the Multi-perturbation Shapley value Analysis (MSA) to uncover the important neurons, as well as the interactions between them, revealing the mechanisms underlying the evolved fault-tolerance. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Tracking Information Flow through the Environment: Simple Cases of Stigmergy

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 563 - 568
      Copyright Year: 2004

      MIT Press eBook Chapters

      Recent work in sensor evolution aims at studying the perception-action loop in a formalized information-theoretic manner. By treating sensors as extracting information and actuators as having the capability to “imprint” information on the environment we can view agents as creating, maintaining and making use of various information flows. In our paper we study the perception-action loop of agents using Shannon information flows. We use information theory to track and reveal the important relationships between agents and their environment. For example, we provide an information-theoretic characterization of stigmergy and evolve finite-state automata as agent controllers to engage in stigmergic communication. Our analysis of the evolved automata and the information flow provides insight into how evolution organizes sensorie information acquisition, implicit internal and external memory, processing and action selection. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Ant Foraging Revisited

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 569 - 574
      Copyright Year: 2004

      MIT Press eBook Chapters

      Most previous artificial ant foraging algorithms have to date relied to some degree on a priori knowledge of the environment, in the form of explicit gradients generated by the nest, by hard-coding the nest location in an easily-discoverable place, or by imbuing the artificial ants with the knowledge of the nest direction. In contrast, the work presented solves ant foraging problems using two pheromones, one applied when searching for food and the other when returning food items to the nest. This replaces the need to use complicated nest-discovery devices with simpler mechanisms based on pheromone information, which in turn reduces the ant system complexity. The resulting algorithm is orthogonal and simple, yet ants are able to establish increasingly efficient trails from the nest to the food in the presence of obstacles. The algorithm replaces the blind addition of new amounts of pheromones with an adjustment mechanism that resembles dynamic programming. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Learning Ant Foraging Behaviors

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 575 - 580
      Copyright Year: 2004

      MIT Press eBook Chapters

      Insects are good at cooperatively solving many complex tasks. For example, foraging for food far away from a nest can be solved through relatively simple behaviors in combination with pheromones. As task complexity increases, however, it may become difficult to find individual agent rules which yield a desired emergent cooperative behavior, or to know if any such rules exist at all. For such tasks, machine learning techniques like evolutionary computation (EC) may prove a valuable approach to searching the space of possible rule combinations. This paper presents an application of genetic programming to search for foraging behaviors. The learned foraging behaviors use only pheromone information to find the path to the nest and to the food source. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Systems Biology Thought Experiments in Human Genetics Using Artificial Life and Grammatical Evolution

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 581 - 586
      Copyright Year: 2004

      MIT Press eBook Chapters

      A goal of systems biology and human genetics is to understand how DNA sequence variations impact human health through a hierarchy of biochemical, metabolic, and physiological systems. We present here a proof-of-principle study that demonstrates how artificial life in the form of agent-based simulation can be used to generate hypothetical systems biology models that are consistent with pre-defined genetic models of disease susceptibility. Here, an evolutionary computing strategy called grammatical evolution is utilized to discover artificial life models. The goal of these studies is to perform thought experiments about the nature of complex biological systems that are consistent with genetic models of disease susceptibility. It is anticipated that the utility of this approach will be the generation of biological hypotheses that can then be tested using experimental systems. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Author Index

      Pollack, J. ; Bedau, M. ; Husbands, P. ; Watson, R. ; Ikegami, T.
      Artificial Life IX:Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems

      Page(s): 587 - 589
      Copyright Year: 2004

      MIT Press eBook Chapters

      Artificial Life is an interdisciplinary effort to investigate the fundamental properties of living systems through the simulation and synthesis of life-like processes. The young field brings a powerful set of tools to the study of how high-level behavior can arise in systems governed by simple rules of interaction. Some of the fundamental questions include:What are the principles of evolution, learning, and growth that can be understood well enough to simulate as an information process?Can robots be built faster and more cheaply by mimicking biology than by the product design process used for automobiles and airplanes?How can we unify theories from dynamical systems, game theory, evolution, computing, geophysics, and cognition?The field has contributed fundamentally to our understanding of life itself through computer models, and has led to novel solutions to complex real-world problems across high technology and human society. This elite biennial meeting has grown from a small workshop in Santa Fe to a major international conference. This ninth volume of the proceedings of the international A-life conference reflects the growing quality and impact of this interdisciplinary scientific community. View full abstract»