Self-learning hybrid fuzzy adaptive genetic algorithm

A novel self-learning hybrid fuzzy genetic algorithm (GA) is proposed to improve the robustness of the adaptive GA (AGA), and to solve more effectively the NP\|hard problem of optimization. In this hybrid algorithm, one special binary-coded standard GA makes it possible for on-line learning the knowledge on AGA . Automatically designing and tuning the fuzzy knowledge-based system, self-learning fuzzy techniques based on GA can find the optimized fuzzy system for AGA by the reinforcement learning method. Simulation experiments show a dynamic parametric AGA system designed by the proposed automatic method, as well as the training results. They also indicate the general applicability of the self-learning hybrid fuzzy AGA to combinatorial optimization similar to the traveling salesman problem.