Inference Algorithm Performance and Selection under Constrained Resources

Abstract : Knowing that reasoning over probabilistic networks is, in general, NP-hard, and that most reasoning environments have limited resources, we need to select algorithms that can solve a given problem as fast as possible. This thesis presents a method for predicting the relative performance of reasoning algorithms based on the domain characteristics of the target knowledge structure. Armed with this knowledge, the research shows how to choose the best algorithm to solve the problem. The effects of incompleteness of the knowledge base at the time of inference is explored, and requirements for reasoning over incompleteness are defined. Two algorithms for reasoning over incomplete knowledge are developed: a genetic algorithm and a best first search. Empirical results indicate that it is possible to predict, based on domain characteristics, which of these algorithms will have better performance on a given problem.

[1]  Gregory F. Cooper,et al.  The Computational Complexity of Probabilistic Inference Using Bayesian Belief Networks , 1990, Artif. Intell..

[2]  D. Wolpert,et al.  No Free Lunch Theorems for Search , 1995 .

[3]  Eric Horvitz,et al.  Reasoning about beliefs and actions under computational resource constraints , 1987, Int. J. Approx. Reason..

[4]  D. E. Goldberg,et al.  Genetic Algorithms in Search, Optimization & Machine Learning , 1989 .

[5]  Shlomo Zilberstein,et al.  Optimal Composition of Real-Time Systems , 1996, Artif. Intell..

[6]  A. E. NicholsonDepartment.  Belief Network Inference Algorithms: a Study of Performance Based on Domain Characterisation , 1996 .

[7]  David Poole,et al.  Average-Case Analysis of a Search Algorithm for Estimating Prior and Posterior Probabilities in Bayesian Networks with Extreme Probabilities , 1993, IJCAI.

[8]  Leslie Pack Kaelbling,et al.  Deliberation Scheduling for Time-Critical Sequential Decision Making , 1993, UAI.

[9]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[10]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1992, Artificial Intelligence.

[11]  Michael Luby,et al.  Approximating Probabilistic Inference in Bayesian Belief Networks is NP-Hard , 1993, Artif. Intell..

[12]  Darwyn O. Banks.  Acquiring Consistent Knowledge for Bayesian Forests. , 1995 .

[13]  Solomon Eyal Shimony,et al.  Finding MAPs for Belief Networks is NP-Hard , 1994, Artif. Intell..

[14]  Solomon Eyal Shimony,et al.  On a Distributed Anytime Architecture for Probabilistic Reasoning. , 1995 .

[15]  Mark S. Boddy,et al.  Solving Time-Dependent Planning Problems , 1989, IJCAI.

[16]  Leslie Pack Kaelbling,et al.  Planning under Time Constraints in Stochastic Domains , 1993, Artif. Intell..

[17]  Mark A. Kramer,et al.  GALGO: A Genetic ALGOrithm Decision Support Tool for Complex Uncertain Systems Modeled with Bayesian Belief Networks , 1993, UAI.

[18]  Eric Horvitz,et al.  Reasoning under Varying and Uncertain Resource Constraints , 1988, AAAI.

[19]  Anas N. Al-Rabadi,et al.  A comparison of modified reconstructability analysis and Ashenhurst‐Curtis decomposition of Boolean functions , 2004 .