– David Goldberg, Genetic Algorithms in Search, Optimization, and.pdf

Evolutionary Algorithms for Optimization in Embedded Systems Neal Bambha nbambha@ ENEE 759E Neal Bambha November 14, 2006 References This lecture is based in part on – Darrell Whitley, “A Genetic Algorithm Tutorial,” Statistics and Computing (4):65–85, 1994 – David Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley, 1989 – Kenneth De Jong, “Evolutionary Computation: A Unified Approach,” 2004 GECCO Tutorial – Eric Goodman, “Introduction to Genetic Algorithms,” 2004 GECCO Tutorial – Franz Rothlauf, “Representations for Evolutionary Algorithms,” 2004 GECCO Tutorial /˜nbambha/enee759e.html ENEE 759E Neal Bambha November 14, 2006 2/30 Evolutionary Algorithms (EAs) Are a method of search, often applied to optimization or learning Are stochastic—but are not random search Use an evolutionary analogy, “survival of the fittest” Global search that does not use gradient information – Non-differentiable functions – Multiple local optima (multimodal) Belong to class of methods known in AI as “weak methods” because they make relatively few assumptions about the problem ENEE 759E Neal Bambha November 14, 2006 3/30 EA vs. Classical Optimization Classical Optimization Evolutionary Algorithm Generates a single point each Generates a population of points iteration. The sequence of points each iteration. The best point in approaches an optimal solution. the population approaches an optimal solution. Selects the next point in the Selects the next population by sequence b