第9章遗传算法.pptVIP

第9章 遗传算法 Genetic Algorithms Evolutionary computation Genetic algorithm (GA) Invented by John Holland 1975 Made popular by John R. Koza 1992 Genetic Programming (GP) “Genetic Programming. On the Programming of Computers by Means of Natural Selection.” - Origin of GP, John R. Koza, Stanford University, 1992 Combining the idea of machine learning and evolved tree structures. Evolutionary computation Evolutionary strategy (ES) Developed: Germany in the 1970’s Early names: I. Rechenberg, H.-P. Schwefel Typically applied to: numerical optimisation Evolutionary Programming (EP) Developed: USA in the 1960’s Early names: D. Fogel Typically applied to: numerical optimisation Simple Example of GA Encode Solution: Just use 5 bits (1 or 0). Generate initial population. Evaluate each solution against objective. Simple Example of GA (cont.） Create next generation of solutions Probability of “being a parent” depends on the fitness. Ways for parents to create next generation Reproduction Use a string again unmodified. Crossover C Cut and paste portions of one string to another. Mutation M Randomly flip a bit. COMBINATION of all of the above. Crossover Chromosom crossover GA crossover Mutation Natural Selection The likelihood of any individual becoming a parent is directly proportional to its relative fitness roulette wheel selection Basic Genetic Algorithm Step 1. Generate a random population of n chromosomes Step 2. Assign a fitness to each individual Step 3. Repeat until n children have been produced Choose 2 parents based on fitness proportional selection Apply genetic operators to copies of the parents Produce new chromosomes home What is GP？ Designs more complex algorithms Starts with What needs to be done Agent figures out How to do it Architecture Design - Hierarchies Produce results that are competitive with human produced results How are Genetic Principles Applied? “Breeding” computer programs. Crossovers. Mutations. Fitness testing. Computer Programs