AIS-BN: An Adaptive Importance Sampling Algorithm for Evidential Reasoning in Large Bayesian Networks

Stochastic sampling algorithms, while an attractive alternative to exact algorithms in very large Bayesian network models, have been observed to perform poorly in evidential reasoning with extremely unlikely evidence. To address this problem, we propose an adaptive importance sampling algorithm, AIS-BN, that shows promising convergence rates even under extreme conditions and seems to outperform the existing sampling algorithms consistently. Three sources of this performance improvement are (1) two heuristics for initialization of the importance function that are based on the theoretical properties of importance sampling in finite-dimensional integrals and the structural advantages of Bayesian networks, (2) a smooth learning method for the importance function, and (3) a dynamic weighting function for combining samples from different stages of the algorithm. We tested the performance of the AIS-BN algorithm along with two state of the art general purpose sampling algorithms, likelihood weighting (Fung & Chang, 1989; Shachter & Peot, 1989) and self-importance sampling (Shachter & Peot, 1989). We used in our tests three large real Bayesian network models available to the scientific community: the CPCS network (Pradhan et al., 1994), the PATHFINDER network (Heckerman, Horvitz, & Nathwani, 1990), and the ANDES network (Conati, Gertner, VanLehn, & Druzdzel, 1997), with evidence as unlikely as 10-41. While the AIS-BN algorithm always performed better than the other two algorithms, in the majority of the test cases it achieved orders of magnitude improvement in precision of the results. Improvement in speed given a desired precision is even more dramatic, although we are unable to report numerical results here, as the other algorithms almost never achieved the precision reached even by the first few iterations of the AIS-BN algorithm.

[1]  Reuven Y. Rubinstein,et al.  Simulation and the Monte Carlo Method , 1981 .

[2]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Anne Lohrli.  Chapman and Hall , 1985 .

[4]  Max Henrion,et al.  Propagating uncertainty in bayesian networks by probabilistic logic sampling , 1986, UAI.

[5]  Judea Pearl,et al.  Fusion, Propagation, and Structuring in Belief Networks , 1986, Artif. Intell..

[6]  Judea Pearl,et al.  Evidential Reasoning Using Stochastic Simulation of Causal Models , 1987, Artif. Intell..

[7]  Max Henrion,et al.  Some Practical Issues in Constructing Belief Networks , 1987, UAI.

[8]  David J. Spiegelhalter,et al.  Local computations with probabilities on graphical structures and their application to expert systems , 1990 .

[9]  Stuart German,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1988 .

[10]  Robert A. Jacobs,et al.  Increased rates of convergence through learning rate adaptation , 1987, Neural Networks.

[11]  Ross D. Shachter,et al.  Simulation Approaches to General Probabilistic Inference on Belief Networks , 2013, UAI.

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

[13]  Kuo-Chu Chang,et al.  Weighing and Integrating Evidence for Stochastic Simulation in Bayesian Networks , 2013, UAI.

[14]  Gregory F. Cooper,et al.  A randomized approximation algorithm for probabilistic inference on bayesian belief networks , 1990, Networks.

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

[16]  Max Henrion,et al.  Search-Based Methods to Bound Diagnostic Probabilities in Very Large Belief Nets , 1991, UAI.

[17]  Gregory F. Cooper,et al.  An Empirical Analysis of Likelihood-Weighting Simulation on a Large, Multiply-Connected Belief Network , 1991, Computers and biomedical research, an international journal.

[18]  D. Heckerman,et al.  ,81. Introduction , 2022 .

[19]  D. Heckerman,et al.  Toward Normative Expert Systems: Part I The Pathfinder Project , 1992, Methods of Information in Medicine.

[20]  Jeremy York,et al.  Use of the Gibbs Sampler in Expert Systems , 1992, Artif. Intell..

[21]  Sampath Srinivas,et al.  A Generalization of the Noisy-Or Model , 1993, UAI.

[22]  Francisco Javier Díez,et al.  Parameter adjustment in Bayes networks. The generalized noisy OR-gate , 1993, UAI.

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

[24]  Mark E. Frisse,et al.  A tutorial introduction to stochastic simulation algorithms for belief networks , 1993, Artif. Intell. Medicine.

[25]  Jean-Louis Golmard,et al.  An algorithm directly finding the K most probable configurations in Bayesian networks , 1994, Int. J. Approx. Reason..

[26]  David Heckerman,et al.  A New Look at Causal Independence , 1994, UAI.

[27]  Gregory M. Provan,et al.  Knowledge Engineering for Large Belief Networks , 1994, UAI.

[28]  Robert M. Fung,et al.  Backward Simulation in Bayesian Networks , 1994, UAI.

[29]  M. Luby,et al.  An Optimal Algorithm for Monte Carlo Estimation (Extended Abstract). , 1995, FOCS 1995.

[30]  Richard M. Karp,et al.  An optimal algorithm for Monte Carlo estimation , 1995, Proceedings of IEEE 36th Annual Foundations of Computer Science.

[31]  Serafín Moral,et al.  Importance sampling algorithms for the propagation of probabilities in belief networks , 1996, Int. J. Approx. Reason..

[32]  Malcolm Pradhan,et al.  Optimal Monte Carlo Estimation of Belief Network Inference , 1996, UAI.

[33]  Ronald L. Wasserstein,et al.  Monte Carlo: Concepts, Algorithms, and Applications , 1997 .

[34]  Cristina Conati,et al.  On-Line Student Modeling for Coached Problem Solving Using Bayesian Networks , 1997 .

[35]  Sylvia Richardson,et al.  Markov Chain Monte Carlo in Practice , 1997 .

[36]  Michael Luby,et al.  An Optimal Approximation Algorithm for Bayesian Inference , 1997, Artif. Intell..

[37]  Serafín Moral,et al.  A Monte Carlo algorithm for probabilistic propagation in belief networks based on importance sampling and stratified simulation techniques , 1998, Int. J. Approx. Reason..

[38]  Michael I. Jordan.  Learning in Graphical Models , 1999, NATO ASI Series.

[39]  Jian Cheng,et al.  Computational Investigation of Low-Discrepancy Sequences in Simulation Algorithms for Bayesian Networks , 2000, UAI.

[40]  Leslie Pack Kaelbling,et al.  Adaptive Importance Sampling for Estimation in Structured Domains , 2000, UAI.

[41]  Jian Cheng,et al.  Latin Hypercube Sampling in Bayesian Networks , 2000, FLAIRS.