Distributed Genetic Tuning Of Fuzzy Rule-Based Systems

The tuning of Fuzzy Rule Base-Systems is necessary to improve their performance after the extraction of rules. This optimization problem can become a hard one when the size of the considered system in terms of the number of variables, rules and data samples is big. To alleviate this growth in complexity, we propose a distributed genetic algorithm which explotes the nowadays available parallel hardware (multicore microprocessors and clusters). The empirical performance in solution quality and computing time is assessed by comparing its results with those from a highly effective sequential tuning algorithm. Both methods are applied for the modeling of four well-known regression problems.