Title :
On learning kDNFns Boolean formulas
Author :
Hernandez-Aquirre, A. ; Buckles, Bill P. ; Coello, Carlos A Coello
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Tulane Univ., New Orleans, LA, USA
Abstract :
The number of samples needed to learn an instance of the representation class kDNFns of Boolean formulas is predicted using some tolerance parameters by the PAC framework. When the learning machine is a simple genetic algorithm, the initial population is an issue. Using PAC-learning we derive the population size that has at least one individual at a given Hamming distance from the optimum. Then we show that the GA evolves solutions from initial populations rather far (Hamming distance) from the optimum
Keywords :
Boolean functions; genetic algorithms; learning (artificial intelligence); Boolean formulas; PAC-learning; genetic algorithm; learning machine; Biological cells; Error correction; Genetic algorithms; Hamming distance; Machine learning; Neural networks; Terminology; Testing; Upper bound;
Conference_Titel :
Evolvable Hardware, 2001. Proceedings. The Third NASA/DoD Workshop on
Conference_Location :
Long Beach, CA
Print_ISBN :
0-7695-1180-5
DOI :
10.1109/EH.2001.937967
