Proceedings of the First International Conference on Genetic Algorithms and their ApplicationsJohn J. Grefenstette Computer solutions to many difficult problems in science and engineering require the use of automatic search methods that consider a large number of possible solutions to the given problems. This book describes recent advances in the theory and practice of one such search method, called Genetic Algorithms. Genetic algorithms are evolutionary search techniques based on principles derived from natural population genetics, and are currently being applied to a variety of difficult problems in science, engineering, and artificial intelligence. |
Contents
Knowledge growth in an artificial animal | |
SESSION 2 | |
The bucket brigade is not genetic | |
synthesis and results | |
Larry A Rendell | |
Improving the performance of genetic algorithms in classifier systems | |
THURSDAY JULY 25 1985 | |
Other editions - View all
Common terms and phrases
Adaptive Algorithms alleles applied approach ARSO average behavior binary bucket brigade classifier system Computer concept consider constraints critical points crossover crossover operator defined detector diffeomorphism differential topology environment equivalent evaluation function example expected value experimental function feature space Figure function optimization GA's gene genetic algorithm genetic operators genotype heuristic hillclimbing Holland hyperplanes image registration implementation improve individuals KL-ONE knowledge structures layout learning classifier system learning system Machine Learning mapping match score message list method mutation object offspring packing parameter pattern recognition payoff percent involvement performance PLS1 PLS2 population position problem properties random randomly rapid convergence recombination region set represent representation reproduction reward scheme rule sample schemata search space selection simple simulation solution specific strategy string task taxa taxon techniques templates Theorem tour traveling salesman problem University of Michigan variables vector VEGA voter