By Topic

Evolutionary Computation, 2003. CEC '03. The 2003 Congress on

8-12 Dec. 2003

Go

Filter Results

Displaying Results 1 - 25 of 84
  • A convergence model for asynchronous parallel genetic algorithms

    Publication Year: 2003, Page(s):2627 - 2634 Vol.4
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1647 KB) | HTML iconHTML

    We describe and verify a convergence model that allows the islands in a parallel genetic algorithm to run at different speeds, and to simulate the effects of communication or machine failure. The model extends on present theory of parallel genetic algorithms and furthermore it provides insight into the design of asynchronous parallel genetic algorithms that work efficiently on volatile and heterog... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An heuristic-based self-adapting crossover method: additional flexibility in the evolutionary process

    Publication Year: 2003, Page(s):2619 - 2626 Vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1663 KB) | HTML iconHTML

    Self-adaptive evolutionary algorithms have gained more attention due to their flexibility to adapt to complex fitness landscape. We present a method to self-adapt crossover parameters of a genetic algorithm during evolution. Not only crossover type but crossover probabilities also are self-adapted allowing the search procedure to find out the most suitable parameters for each search phase. A new h... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Evolutionary multi-objective optimisation with a hybrid representation

    Publication Year: 2003, Page(s):2262 - 2269 Vol.4
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1622 KB) | HTML iconHTML

    For tackling multiobjective optimisation (MOO) problem, many methods are available in the field of evolutionary computation (EC). To use the proposed method(s), the choice of the representation should be considered first. In EC, often binary representation and real-valued representation are used. We propose a hybrid representation, composed of binary and real-valued representations for multi-objec... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the connection between the no free lunch theorem and the trivial property for recursively enumerable languages

    Publication Year: 2003, Page(s):2611 - 2618 Vol.4
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1686 KB) | HTML iconHTML

    We return to the no free lunch theorem, which is one of the most important theorems from the evolutionary computation foundations. We show that the no free lunch theorem can be interpreted as a trivial property of recursively enumerable languages. We demonstrate that if we consider not all problems and cost functions, i.e., a nontrivial property, the problem of finding the best evolutionary algori... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Effects of repair procedures on the performance of EMO algorithms for multiobjective 0/1 knapsack problems

    Publication Year: 2003, Page(s):2254 - 2261 Vol.4
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1630 KB) | HTML iconHTML

    Multiobjective 0/1 knapsack problems have been used for examining the performance of EMO (evolutionary multiobjective optimization) algorithms in the literature. We demonstrate that their performance on such a test problem strongly depends on the choice of a repair procedure. We show through computational experiments that much better results are obtained from greedy repair based on a weighted scal... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The simple supply chain model and evolutionary computation

    Publication Year: 2003, Page(s):2322 - 2329 Vol.4
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1666 KB) | HTML iconHTML

    This article provides an overview of the simple supply chain model (SSCM), scenarios derived from this model and the strategies being used to begin to tackle SSCM problems. The article further provides details of how evolutionary computation is used (via population based incremental learning) to optimise parameters for the designed strategies. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Emergence of adaptive behaviors by redundant robots - robustness to changes environment and failures

    Publication Year: 2003, Page(s):2572 - 2579 Vol.4
    Cited by:  Papers (3)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1657 KB) | HTML iconHTML

    Acquiring adaptive behaviors of robots automatically is one of the most interesting topics of the evolutionary systems. In previous works, we have developed an adaptive autonomous control method for redundant robots. The QDSEGA is one of the methods that we have proposed for them. The QDSEGA is realized by combining Q-learning and GA, and it can acquire suitable behaviors by adapting a movement of... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Asynchronous parallel distributed GA using elite server

    Publication Year: 2003, Page(s):2603 - 2610 Vol.4
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1637 KB) | HTML iconHTML

    To speed up GA search, parallel distributed genetic algorithms are used. However in the current asynchronous parallel distributed genetic algorithm like the random-exchange or the sigma-exchange, it is hard to implement on the parallel computers or on the WS/PC clusters on the network, and it is easy to deadlock. We introduce an implementation method for asynchronous parallel distributed genetic a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The evolution of blackjack strategies

    Publication Year: 2003, Page(s):2474 - 2481 Vol.4
    Cited by:  Papers (5)  |  Patents (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1775 KB) | HTML iconHTML

    We investigate the evolution of a blackjack player. We utilise three neural networks (one for splitting, one for doubling down and one for standing/hitting) to evolve blackjack strategies. Initially a pool of randomly generated players play 1000 hands of blackjack. An evolutionary strategy is used to mutate the best networks (with the worst networks being killed). We compare the best evolved strat... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Financial forecasting through unsupervised clustering and evolutionary trained neural networks

    Publication Year: 2003, Page(s):2314 - 2321 Vol.4
    Cited by:  Papers (8)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1675 KB) | HTML iconHTML

    We present a time series forecasting methodology and applies it to generate one-step-ahead predictions for two daily foreign exchange spot rate time series. The methodology draws from the disciplines of chaotic time series analysis, clustering, artificial neural networks and evolutionary computation. In brief, clustering is applied to identify neighborhoods in the reconstructed state space of the ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A fast algorithm on finding the non-dominated set in multi-objective optimization

    Publication Year: 2003, Page(s):2565 - 2571 Vol.4
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1508 KB) | HTML iconHTML

    A fast algorithm is proposed to find the nondominated set for multiobjective optimization problems in this paper. Two accelerated techniques are adopted in the algorithm. One is that the algorithm can yield an integer rank set after it indexes the search space. Based on this, the goal is changed into determination of the nondominated set of the integer rank set. The other is that the nondominated ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multiple single objective Pareto sampling

    Publication Year: 2003, Page(s):2678 - 2684 Vol.4
    Cited by:  Papers (62)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1572 KB) | HTML iconHTML

    We detail a new nonPareto evolutionary multiobjective algorithm, multiple single objective Pareto sampling (MSOPS), that performs a parallel search of multiple conventional target vector based optimisations, e.g. weighted min-max. The method can be used to generate the Pareto set and analyse problems with large numbers of objectives. The method allows bounds and discontinuities of the Pareto set t... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improving genetic classifiers with a boosting algorithm

    Publication Year: 2003, Page(s):2596 - 2602 Vol.4
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1444 KB) | HTML iconHTML

    We present a boosting genetic algorithm for classification rule discovery. The method is based on the iterative rule learning approach to genetic classifiers. The boosting mechanism increases the weight of those training instances that are not classified correctly by the new rules, so that in the next iteration the algorithm focuses the search on those rules that capture the misclassified or uncov... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Diversity control in a multi-objective genetic algorithm

    Publication Year: 2003, Page(s):2704 - 2711 Vol.4
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1640 KB) | HTML iconHTML

    We cover an investigation on the effects of diversity control in a multiobjective genetic algorithm (MOGA). Specifically, the diversity control operator used is based on the one developed for a diversity control oriented genetic algorithm (DCGA). The performance comparison between multiobjective genetic algorithms with and without diversity control is explored where different benchmark problems wi... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Piece difference: simple to evolve?

    Publication Year: 2003, Page(s):2470 - 2473 Vol.4
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1380 KB) | HTML iconHTML

    We detail a study into whether the 'piece difference' heuristic could be evolved for the game of checkers. A co-evolutionary algorithm is used to evolve a piece-weighting system that is used as the evaluation function in a minimax checkers player. The results suggest that the 'piece difference' heuristic will evolve if allowed, but it is not necessarily easy. The work has also demonstrated that ot... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of immunological memory to the color classification of tiles

    Publication Year: 2003, Page(s):2815 - 2820 Vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1430 KB) | HTML iconHTML

    Based on mechanism that the vertebrate immune system remembers the antigen it has met before by retaining in the body some memory cells, an algorithm is proposed to search for the representative of the data set by generating its memory cells. The algorithm is first tested on a two-dimensional data set with three cluster centers to see if the memory cells built could really be representative. Then ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A co-evolutionary approach to the tacit collusion of generators in oligopolistic electricity markets: piecewise linear bidding structure case

    Publication Year: 2003, Page(s):2306 - 2313 Vol.4
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1682 KB) | HTML iconHTML

    Wholesale electricity markets now operate in many countries around the world. These markets determine a spot price for electricity as the clearing price when generators bid in power at various prices. Because these markets are by nature repeated day after day, they are prone to tacit collusion between generators to raise prices. Tacit collusion occurs when a generator could improve its profits by ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Two mode Q-learning

    Publication Year: 2003, Page(s):2449 - 2454 Vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1519 KB) | HTML iconHTML

    In this paper, a new two mode Q-learning using both the success and failure experiences of an agent is proposed for the fast convergence, which extends Q-learning, a well-known scheme used for reinforcement learning. In the Q-learning, if the agent enters into the "fail" state, it receives a punishment from environment. By this punishment, the Q value of the action which generated the failure expe... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A distributed cooperative coevolutionary algorithm for multiobjective optimization

    Publication Year: 2003, Page(s):2513 - 2520 Vol.4
    Cited by:  Papers (6)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1650 KB) | HTML iconHTML

    Evolutionary techniques have become one of the most powerful tools for solving multiobjective optimization (MOO) problems. However the computational cost involved in terms of time and hardware often become surprisingly burdensome as the size and complexity of the problem increases. We propose a distributed cooperative coevolutionary algorithm (DCCEA), which evolves multiple solutions in the form o... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Evolving adaptive neural networks with and without adaptive synapses

    Publication Year: 2003, Page(s):2557 - 2564 Vol.4
    Cited by:  Papers (18)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1686 KB) | HTML iconHTML

    A potentially powerful application of evolutionary computation (EC) is to evolve neural networks for automated control tasks. However, in such tasks environments can be unpredictable and fixed control policies may fail when conditions suddenly change. Thus, there is a need to evolve neural networks that can adapt, i.e. change their control policy dynamically as conditions change. In this paper, we... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Self-optimizing fuzzy controller based on extreme evolution algorithm

    Publication Year: 2003, Page(s):2673 - 2677 Vol.4
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1475 KB) | HTML iconHTML

    Extreme evolution algorithm (EEA) is presented to solve fast global optimization problems. The algorithm selects parents according to extreme law but not to the fitness law. Recombining extreme elements obviously accelerates evolution procedure. Secondly, we construct a self-optimizing fuzzy controller based on the EEA. The controller shows a good performance on nonlinear optimization control. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neuroevolution for reinforcement learning using evolution strategies

    Publication Year: 2003, Page(s):2588 - 2595 Vol.4
    Cited by:  Papers (15)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1647 KB) | HTML iconHTML

    We apply the CMA-ES, an evolution strategy which efficiently adapts the covariance matrix of the mutation distribution, to the optimization of the weights of neural networks for solving reinforcement learning problems. It turns out that the topology of the networks considerably influences the time to find a suitable control strategy. Still, our results with fixed network topologies are significant... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Differential evolution for multi-objective optimization

    Publication Year: 2003, Page(s):2696 - 2703 Vol.4
    Cited by:  Papers (30)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1624 KB) | HTML iconHTML

    Two test problems on multiobjective optimization (one simple general problem and the second one on an engineering application of cantilever design problem) are solved using differential evolution (DE). DE is a population based search algorithm, which is an improved version of genetic algorithm (GA), Simulations carried out involved solving (1) both the problems using Penalty function method, and (... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Molecular dynamics simulation using coarse-grained model to study protein function and beyond

    Publication Year: 2003, Page(s):2719 - 2726 Vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1678 KB) | HTML iconHTML

    To make an inquiry into the mechanisms of biomolecular functions, particularly of protein molecule's, we conducted molecular dynamics (MD) simulations of protein molecules using coarse-grained models that preserve 3-dimensional native structure information, together with the hope of inquiring dynamical aspect of design principle of not only molecular machines but also biomolecular computing system... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Future View: Web navigation based on learning user's browsing patterns by classifier systems

    Publication Year: 2003, Page(s):2829 - 2836 Vol.4
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (1661 KB) | HTML iconHTML

    In this paper, we propose a Future View system that assists user's usual Web browsing. A Future View prefetches Web pages based on user's browsing strategies and present them to a user in order to assist Web browsing. To learn browsing patterns for a user, Future View uses two types of learning classifier systems: a content-based classifier system for contents change patterns and an action-based c... View full abstract»

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