By Topic

Systems, Man and Cybernetics, IEEE Transactions on

Issue 11 • Date Nov 1994

Filter Results

Displaying Results 1 - 6 of 6
  • Dynamic belief networks for discrete monitoring

    Page(s): 1593 - 1610
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1676 KB)  

    We describe the development of a monitoring system which uses sensor observation data about discrete events to construct dynamically a probabilistic model of the world. This model is a Bayesian network incorporating temporal aspects, which we call a dynamic belief network; it is used to reason under uncertainty about both the causes and consequences of the events being monitored. The basic dynamic construction of the network is data-driven. However the model construction process combines sensor data about events with externally provided information about agents' behavior, and knowledge already contained within the model, to control the size and complexity of the network. This means that both the network structure within a time interval, and the amount of history and detail maintained, can vary over time. We illustrate the system with the example domain of monitoring robot vehicles and people in a restricted dynamic environment using light-beam sensor data. In addition to presenting a generic network structure for monitoring domains, we describe the use of more complex network structures which address two specific monitoring problems, sensor validation and the data association problem View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Graph-grammar assistance for automated generation of influence diagrams

    Page(s): 1625 - 1642
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1860 KB)  

    One of the most difficult aspects of modeling complex dilemmas in decision-analytic terms is composing a diagram of relevance relations from a set of domain concepts. Decision models in many domains, however, exhibit certain prototypical patterns that can guide the modeling process. Concepts can be classified according to semantic types that have characteristic positions and typical roles in an influence-diagram model. The authors have developed a graph-grammar production system that uses such inherent interrelationships among terms to facilitate the modeling of medical decisions. The authors' system also can examine a set of graph-grammar rules to establish whether the grammar satisfies a number of properties that they have determined to be important in the derivation of influence-diagram models. The authors' findings suggest that syntactic patterns can lead to automated construction of decision models in domains other than medicine View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamic construction and refinement of utility-based categorization models

    Page(s): 1653 - 1663
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1100 KB)  

    The actions taken by an automated decision-making agent can be enhanced by including mechanisms that enable the agent to categorize concepts effectively. We pose a utility-based approach to categorization based on the idea that categorization should be carried out in the service of action. The choice of concepts is critical in the effective selection of actions under resource constraints. We propose a decision-theoretic framework for categorization which involves reasoning about alternative categorization models consisting of sets of interrelated concepts at varying levels of abstraction. Categorization models that are too abstract may overlook details that are critical for selecting the most appropriate actions. Categorization models that are too detailed, however, may be too expensive to process and may contain irrelevant information. Categorization models are therefore evaluated on the basis of the expected value of their recommended action, taking into account the resource cost of their evaluation. A knowledge representation scheme, known as probabilistic conceptual networks, has been developed to support the dynamic construction of models at varying levels of abstraction. This scheme combines the formalisms of influence diagrams from decision analysis and inheritance/abstraction hierarchies from AI. We also propose an incremental approach to categorical reasoning. By applying decision-theoretic control of model refinement, a resource-constrained actor iteratively decides between continuing to improve the current level of abstraction in the model, or to act immediately View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Tradeoffs in knowledge-based construction of probabilistic models

    Page(s): 1580 - 1592
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1336 KB)  

    In many domains, the ability to use a knowledge base to automatically construct alternative probabilistic network models and then compare them is desirable. This paper makes two novel contributions towards achieving that goal: first, it analyzes a parameterized class of (a) static, and (b) temporal influence diagram, models which differ in the time-series process describing the temporal evolution of the system being modeled. Second, it applies general scoring metrics for comparing these models with respect to predictive accuracy and computational efficiency. The network rankings facilitate comparing the accuracy/efficiency tradeoffs entailed in using TIDs which differ in (1) the accuracy of capturing the temporal evolution of a dynamic system and (2) data and computational requirements. The scoring metrics are used to compare networks in which all variables evolve according to a Markov process with two novel domain-dependent network approximations. These approximations model the evolution of a parsimonious subset of variables rather than all variables View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Inference-driven construction of valuation systems from first-order clauses

    Page(s): 1611 - 1624
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1344 KB)  

    Network-based representations of uncertain knowledge are inherently propositional, and cannot easily accommodate generic, problem independent knowledge. A tremendous effort has been devoted in the knowledge representation community to develop languages that adequately represent different types of generic knowledge. The authors propose inference-driven construction as a means to use these languages for extending the expressiveness of uncertainty networks: they let a knowledge representation system represent generic knowledge and infer solutions to specific problem instances, and then copy the resulting inference structure to an uncertainty network that models these instances. From a dual perspective, inference-driven construction is a way of extending an existing knowledge representation system by attaching an uncertainty calculus to it. In this paper, the authors focus on one particular case of inference-driven construction: building Shenoy-Shafer's valuation systems from uncertain knowledge expressed in the form of first order clauses annotated by Dempster-Shafer's measures of belief. The authors detail an automatic construction procedure for this case, discuss a sample implementation, and provide a soundness and completeness result View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Metareasoning and the problem of small worlds

    Page(s): 1643 - 1652
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1048 KB)  

    Practical decision theoretic reasoning requires the construction of local, problem-specific models in which attention is confined to a restricted universe of propositions. Such a restricted universe is called a small world. Managing the construction and revision of small world models can itself be viewed as a meta-level decision problem. This paper presents a theoretical framework for understanding and analyzing many of the issues associated with the management of small world models View full abstract»

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