×

Directing genetic algorithms for probabilistic reasoning through reinforcement learning. (English) Zbl 1113.68543

Summary: We develop an efficient online approach for belief revision over Bayesian networks by using a reinforcement learning controller to direct a genetic algorithm. The random variables of a Bayesian network can be grouped into several sets reflecting the strong probabilistic correlations between random variables in the group. We build a reinforcement learning controller to identify these groups and recommend the use of “group” mutation and “group” crossover for the genetic algorithm based on these groupings online. The system then evaluates the performance of the genetic algorithm based on these groupings online. The system then evaluates the performance of the genetic algorithm and continues with reinforcement learning to further tune the controller to search for a better grouping.

MSC:

68T37 Reasoning under uncertainty in the context of artificial intelligence
68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1016/0004-3702(90)90060-D · Zbl 0717.68080 · doi:10.1016/0004-3702(90)90060-D
[2] Dietterich T. G., AI Magazine 18 pp 97– (1997)
[3] Kaelbling L. P., Journal of Artificial Intelligence Research 4 pp 237– (1997)
[4] Rojas-Guzman Carlos, Proceedings of the Conference on Uncertainty pp 368– (1993)
[5] Santharam G., IEEE Transactions on System, Man and Cybernetics-PartA: Systems and Humans 72 pp 558– (1997)
[6] Santos Eugene, IEEE Transactions 28 (4) pp 377– (1998)
[7] DOI: 10.1016/S0888-613X(97)00012-1 · Zbl 0939.68119 · doi:10.1016/S0888-613X(97)00012-1
[8] DOI: 10.1016/0004-3702(94)90072-8 · Zbl 0818.68097 · doi:10.1016/0004-3702(94)90072-8
[9] Williams Edward, Proceedings of the AAAI Workshop on Building Resource-Bounded pp 86– (1997)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.