Proceedings of the Seventh Conference of the Society for the Study of Artificial Intelligence and Simulation of Behaviour |
From inside the book
Results 1-3 of 10
Page 139
... population of strings must exist . Genetic algorithms start by ran- domly generating this population . The strings are then evaluated to obtain a quantitative measure of how well they perform as possible problem solutions . Reproduc ...
... population of strings must exist . Genetic algorithms start by ran- domly generating this population . The strings are then evaluated to obtain a quantitative measure of how well they perform as possible problem solutions . Reproduc ...
Page 142
... population of 200 was 0.006 and the total error for all 50 runs was about 0.03 . The 424 - encoder took about 20 times more recom- binations to optimize using a population of 200 geno- types . Again , the algorithms were ran 50 times ...
... population of 200 was 0.006 and the total error for all 50 runs was about 0.03 . The 424 - encoder took about 20 times more recom- binations to optimize using a population of 200 geno- types . Again , the algorithms were ran 50 times ...
Page 143
... population , the more progress can be made before the population converges . Increas- ing the population size translates into more genetic diversity which increases the amount of genetic material available to the search . The consistent ...
... population , the more progress can be made before the population converges . Increas- ing the population size translates into more genetic diversity which increases the amount of genetic material available to the search . The consistent ...
Other editions - View all
Common terms and phrases
abstract action agent analogy Artificial Intelligence assumption axioms behaviour belief circumscription clauses Cognitive complete complex nodes Computer concept condition consistent constraint satisfaction problem constraints constructive negation context model counterfactual DATR decision tree declarative program default inferences defined definition DePlan domain dynamic environment example Figure forced paths formulas function gazing genetic algorithm given graph heuristics higher-level goals implementation instance instant coffee interval k-consistency Kahney knowledge base label Logic Programming mapping match minimal negation as failure neural nextstate notion operators paper partial evaluation population preconditions predicate preference rule problem procedural program procedure proof propagation proposition prototype theory range reasoning recursion relations relationships representation represented robot satisfies schema semantics simulation sinusoidal solving specific state-space structure subgoals suppressed premises symbols task theorem theory extension tion University of Sussex values variables worldstate