Learning Disjunctive Concepts by Means of Genetic Algorithms
作者： Attilio Giordana , Lorenza Saitta , Floriano Zini
DOI: 10.1016/B978-1-55860-335-6.50020-9
关键词:
摘要: REGAL is a Distributed Genetic Algorithm designed for learning concept descriptions from examples, in First Order Logic. In particular, each individual the population represents conjunctive formula VL2 language. order to increase efficiency of generalization process, has been provided with new selection operator, called Universal Suffrage which guarantees (in probability) maintain covering all events. As mostly takes place when two individuals different sets examples are crossed, global capability system increased. Moreover, case disjunctive or multiple concepts, universal suffrage algorithm allows formation species, one corresponding disjunct. this way, disjuncts can be learned parallel obtaining, average, more general solutions than by them at time. A formal analysis operator presented, providing theoretical explanations experimentally observed behaviour. comparison classical and sharing function method also made. Finally, long term control strategy, “Tories Whigs”, proposed overcome problem lethal matings between uncompatible disjuncts. The effectiveness demonstrated on several problems.
elsevier.com
sci-hub.st 参考文章(9)
Gilles Venturini, SIA: a supervised inductive algorithm with genetic search for learning attributes based concepts european conference on machine learning. pp. 280- 296 ,(1993) , 10.1007/3-540-56602-3_142
Cezary Z. Janikow, A New System for Inductive Learning in Attribute-Based Spaces international syposium on methodologies for intelligent systems. pp. 378- 388 ,(1991) , 10.1007/3-540-54563-8_101
David E. Goldberg, Sizing Populations for Serial and Parallel Genetic Algorithms international conference on genetic algorithms. pp. 70- 79 ,(1989)
Michael R. Leuze, Chrisila C. Pettey, A theoretical investigation of a parallel genetic algorithm international conference on genetic algorithms. pp. 398- 405 ,(1989)
Ryszard S. Michalski, A theory and methodology of inductive learning Computer Compacts. ,vol. 1, pp. 49- ,(1983) , 10.1016/0167-7136(83)90132-4
David Perry Greene, Stephen F. Smith, Competition-Based Induction of Decision Models from Examples Machine Learning. ,vol. 13, pp. 229- 257 ,(1993) , 10.1007/BF00993044
Cezary Z. Janikow, A Knowledge-Intensive Genetic Algorithm for Supervised Learning Machine Learning. ,vol. 13, pp. 189- 228 ,(1993) , 10.1007/BF00993043
David E. Goldberg, Jon Richardson, Genetic algorithms with sharing for multimodal function optimization international conference on genetic algorithms. pp. 41- 49 ,(1987)
J. H. Holland, Escaping Brittleness: The Possibilities of General-Purpose Learning Algorithms Applied to Parallel Rule-Based Systems Machine Learning: An Artificial Intelligence Approach. ,vol. 2, pp. 593- 623 ,(1986)