Function decomposition in machine learning
作者： Blaž Zupan , Ivan Bratko , Marko Bohanec , Janez Demšar
关键词: Chess endgame 、 Computer science 、 Hierarchy 、 Functional decomposition 、 Artificial intelligence 、 Problem domain 、 Boolean function 、 Computational learning theory 、 Wake-sleep algorithm 、 Feature learning 、 Classification rule 、 Machine learning 、 Conceptual graph 、 Hierarchy (mathematics)
摘要: To solve a complex problem, one of the effective general approaches is to decompose it into smaller, less and more manageable subproblems. In machine learning, this principle foundation for structured induction [44]: instead learning single classification rule from examples, define concept hierarchy learn rules each (sub)concepts. Shapiro [44] used fairly chess endgame demonstrated that complexity comprehensiveness (“brain-compatibility”) obtained solution was superior unstructured one. helped by master structure his problem domain. Typically, applications involve manual development selection examples induce subconcept rules; usually tiresome process requires an active availability domain expert over long periods time. Therefore, would be very desirable automate decomposition task.
springer.com
sci-hub.st 参考文章(50)
Stuart J. Russell, Tree-structured bias national conference on artificial intelligence. pp. 641- 645 ,(1988)
Marko Bohanec, DEX: An Expert System Shell for Decision Support • ,(1990)
Igor Mozetic, Learning of qualitative models EWSL'87 Proceedings of the 2nd European Conference on European Working Session on Learning. pp. 201- 217 ,(1987)
Marko Bohanec, Ivan Bratko, Janez Demsar, Blaz Zupan, Induction of Concept Hierarchies from Noisy Data international conference on machine learning. pp. 1199- 1206 ,(2000)
Stephen Muggleton, Structuring knowledge by asking questions EWSL'87 Proceedings of the 2nd European Conference on European Working Session on Learning. pp. 218- 229 ,(1987)
Stephen H. Muggleton, Inductive acquisition of expert knowledge ,(1989)
Steven L. Salzberg, Alberto Segre, Programs for Machine Learning ,(1994)
Donald Michie, Problem Decomposition and the Learning of Skills european conference on machine learning. pp. 17- 31 ,(1995) , 10.1007/3-540-59286-5_46
Ron Kohavi, Bottom-up induction of oblivious read-once decision graphs european conference on machine learning. pp. 154- 169 ,(1994) , 10.1007/3-540-57868-4_56
A. Shapiro, T. Niblett, AUTOMATIC INDUCTION OF CLASSIFICATION RULES FOR A CHESS ENDGAME Advances in Computer Chess. pp. 73- 92 ,(1982) , 10.1016/B978-0-08-026898-9.50010-3