ion Modulation In many cases, it may be more useful to do normative inference on a model that is deemed to be complete at a particular level of abstraction than it is to do an approximate or heuristic analysis of a model that is too large to be analyzed under specific resource constraints. It may prove useful in many cases to store several belief-network representations, each containing propositions at different levels of abstraction. In many domains, models at higher levels of abstraction are more tractable. As the time available for computation decreases, network modules of increasing abstraction can be employed. ffl Local Reformulation Local reformulation is the modification of specific troublesome topologies in a belief network. Approximation methods and heuristics designed to modify the microstructure of belief networks will undoubtedly be useful in the tractable solution of large uncertain-reasoning problems. Such strategies might be best applied at knowledge-encoding time. An...

Despite their different perspectives, artificial intelligence (AI) and the disciplines of decision science have common roots and strive for similar goals. This paper surveys the potential for addressing problems in representation, inference, knowledge engineering, and explanation within the decision-theoretic framework. Recent analyses of the restrictions of several traditional AI reasoning techniques, coupled with the development of more tractable and expressive decisiontheoretic representation and inference strategies, have stimulated renewed interest in decision theory and decision analysis. We describe early experience with simple probabilistic schemes for automated reasoning, review the dominant expert-system paradigm, and survey some recent research at the crossroads of AI and decision science. In particular, we present the belief network and influence diagram representations. Finally, we discuss issues that have not been studied in detail within the expert-systems sett...

Pathfinder is an expert system that assists surgical pathologists with the diagnosis of lymph-node diseases. The program is one of a growing number of normative expert systems that use probability and decision theory to acquire, represent, manipulate, and explain uncertain medical knowledge. In this article, we describe Pathfinder and our research in uncertain-reasoning paradigms that was stimulated by the development of the program. We discuss limitations with early decision-theoretic methods for reasoning under uncertainty and our initial attempts to use non-decision-theoretic methods. Then, we describe experimental and theoretical results that directed us to return to reasoning methods based in probability and decision theory.

We address practical issues concerning the construction and use of decision-theoretic or normative expert systems for diagnosis. In particular, we examine Pathfinder, a normative expert system that assists surgical pathologists with the diagnosis of lymphnode diseases, and discuss the representation of dependencies among pieces of evidence within this system. We describe the belief network, a graphical representation of probabilistic dependencies. We see how Pathfinder uses a belief network to construct differential diagnoses efficiently, even when there are dependencies among pieces of evidence. In addition, we introduce an extension of the belief-network representation called a similarity network, a tool for constructing large and complex belief networks. The representation allows a user to construct independent belief networks for subsets of a given domain. A valid belief network for the entire domain can then be constructed from the individual belief networks. We also introduce the partition, a graphical representation that facilitates the assessment of probabilities associated with a belief network. Finally, we show that the similarity-network and partition representations made practical the construction of Pathfinder.

As Bayesian networks are applied to more complex and realistic real-world applications, the development of more efficient inference algorithms working under real-time constraints is becoming more and more important. This paper presents a survey of various exact and approximate Bayesian network inference algorithms. In particular, previous research on real-time inference is reviewed. It provides a framework for understanding these algorithms and the relationships between them. Some important issues in real-time Bayesian networks inference are also discussed.

Monte Carlo sampling has become a major vehicle for approximate inference in Bayesian networks. In this paper, we investigate a family of related simulation approaches, known collectively as quasi-Monte Carlo methods based on deterministic low-discrepancy sequences. We first

This paper describes an algorithmic means for inducing implication networks from empirical data samples. The induced network enables efficient inferences about the values of network nodes if certain observations are made. This implication induction method is approximate in nature as probablistic network requirements are relaxed in the construction of dependence relationships based on statistical testing. In order to examine the effectiveness and validity of the induction method, several Monte-Carlo simulations were conducted where theoretical Bayesian networks were used to generate empirical data samples \Gamma some of which were used to induce implication relations whereas others were used to verify the results of evidential reasoning with the induced networks. The values in the implication networks were predicted by applying a modified version of Dempster-Shafer belief updating scheme. The results of predictions were, furthermore, compared to the ones generated by Pearl's stochastic ...

Researchers in artificial intelligence and decision analysis share a concern with the construction of formal models of human knowledge and expertise. Historically, however, their approaches to these problems have diverged. Members of these two communities have recently discovered common ground: a family of graphical models of decision theory known as influence diagrams or as belief networks. These models are equally attractive to theoreticians, decision modelers, and designers of knowledge-based systems. From a theoretical perspective, they combine graph theory, probability theory and decision theory. From an implementation perspective, they lead to powerful automated systems. Although many practicing decision analysts have already adopted influence diagrams as modeling and structuring tools, they may remain unaware of the theoretical work that has emerged from the artificial intelligence community. This paper surveys the first decade or so of this work. Investment Technology Group, ...

ion in Models The full probabilistic and decision analytic framework for reasoning about uncertainty is very attractive and has a long history of advocacy and analysis (e.g., [17,30]). Nevertheless, applying these ideas in a straightforward manner requires accurate elicitation of many numeric probabilities and utilities. The difficulty of doing this begs for practical or conceptual simplification. Early AIM and AI programs introduced a large variety of scoring schemes that were thought, at the time, to be simpler or more attractive than probability theory. In retrospect, however, many of these have been shown to be equivalent to standard probability theory, with perhaps a few additional assumptions or approximations (e.g., [12] concerning Mycin, and the discussion of log likelihood ratios, above, for Internist). Some of the schemes were originally introduced simply because the methods of Bayes networks were unknown, yet the need for chains of probabilistic inference was critical (e.g...

This paper presents Bayesian networks as a framework for distributed reasoning in expert systems. We discuss methods for evidence propagation, for learning, with emphasis on sequential learning, and for generating linguistic explanations. When a parallel implementation is possible, we describe the computational power, i.e. the information that must be stored and the local calculations that must be performed at every node, in order to get a distributed expert system. Finally, a brief comparison to neural networks is offered.