Population-Based Incremental Learning: A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning

摘要: Genetic algorithms (GAs) are biologically motivated adaptive systems which have been used, with varying degrees of success, for function optimization. In this study, an abstraction the basic genetic algorithm, Equilibrium Algorithm (EGA), and GA in turn, reconsidered within framework competitive learning. This new perspective reveals a number different possibilities performance improvements. paper explores population-based incremental learning (PBIL), method combining mechanisms generational algorithm simple The combination these two methods tool is far simpler than GA, out-performs on large set optimization problems terms both speed accuracy. presents empirical analysis where proposed technique will outperform algorithms, describes class may be able to perform better. Extensions discussed analyzed. PBIL extensions compared standard twelve problems, including numerical functions, traditional test suite NP-Complete problems.