Evolutionary algorithm for learning Bayesian structures from data

• Autorzy: Kozłowski M. , Wierzchoń S. T.
• Konferencja: 8th Workshop of the Polish Society of Computer Simulation (30.08-1.09.2001, Gdańsk-Sobieszewo, Poland)
• Języki publikacji: EN

Abstrakty

EN

In this paper we report an evolutionary approach to learning Bayesian networks from data. We explain reasons, which advocate such a non-deterministic approach. We analyze weaknesses of previous works and come to conclusion that we should operate in the search space native for the problem i.e. in the space of directed acyclic graphs instead of standard space of binary strings. This requires adaptation of evolutionary methodology into very specific needs. We propose quite new data representation and implementation of generalized genetic operators and then we present an efficient algorithm capable of learning complex networks without additional assumptions. We discuss results obtained with this algorithm. The approach presented in this paper can be extended with the possibility to absorb some suggestions from experts or obtained by means of data preprocessing.

Słowa kluczowe

EN

Bayesian networks; structure learning; evolutionary algorithm; discrete optimization

• Wydawca: Politechnika Gdańska
• Czasopismo: TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk
• Rocznik: 2002
• Tom: Vol. 6, No 3
• Strony: 509--521
• Opis fizyczny: Bibliogr. 15 poz., rys., tab.

Twórcy

autor

Kozłowski M.

• Faculty of Mathematics and Information Science, Warsaw University of Technology, Plac Politechniki 1, 00-661 Warsaw, Poland

autor

Wierzchoń S. T.

• Institute of Computer Science, Polish Academy of Sciences, Ordona 21, 01-267 Warsaw, Poland
• Department of Computer Science, Technical University of Bialystok, Wiejska 45a, 15-333 Bialystok, Poland

Bibliografia

• [1] Jordan M I (Ed.) 1998 Learning in Graphical Models, The MIT Press, London
• [2] Lauritzen S L 1996 Graphical Models, Oxford University Press
• [3] Pearl J 1986 Artificial Intelligence 29 241
• [4] Heckerman D in [1] pp. 301-354
• [5] Cowell R G, David A P, Lauritzen S L and Spegelhalter D J 1999 Probabilistic Networks and Expert Systems, Springer-Verlag, New York
• [6] Robinson R W 1997 Lecture Notes in Mathematics 622: Combinatorial Mathematics V (Little CHC, Ed.), Springer-Verlag, pp. 28-43
• [7] Chickering D M, Geiger D and Heckerman D 1994 Technical Report MSR-TR-94-17, Redmond WA: Microsoft Research
• [8] Cooper G F and Herskovits E 1992 Machine Learning 9 309
• [9] Buntine W 1996 IEEE Trans, on Knowledge and Data Engineering 8 195
• [10] Geiger D and Heckerman D 1995 Uncertainty in Artificial Intelligence 11 (Besnard P and Hanks S, Eds.), Morgan Kaufmann, pp. 196-207
• [11] Larranaga P, Poza M, Yurramendi Y, Murga R H and Kuijpers C M H 1996 IEEE Trans, on Pattern Analysis and Machine Intelligence 18 912
• [12] Goldberg D E 1989 Genetic Algorithms in Search, Optimization and Machine Learning, Reading, Mass.: Addison-Wesley
• [13] Beinlich I A, Suermondt H J, Chavez R M and Cooper G F 1989 Proc. 2nd European Conf. on Artificial Intelligence in Medicine, London, Springer-Verlag, pp. 247-256
• [14] Eigen m 1971 Naturwissenschaften 58 465
• [15] Peliti L Introduction to the Statistical Theory of Darwinian Evolution, arXiv:cond-mat/9712027