Combination of online Clustering and Q-value based genetic reinforcement learning for fuzzy network design
Chia‐Feng Juang, Chunfeng Lu · 2004
This paper proposes a combination of on-line Clustering and Q-value based Genetic Algorithm learning scheme for Fuzzy network design (CQGAF) with reinforcements. The CQGAF fulfills GA based fuzzy network design under reinforcement learning environment where only weak reinforcement signals such as "success" and "failure" are available. In CQGAF, there are no fuzzy rules initially. They are generated automatically. The precondition part of a fuzzy network is online constructed by an aligned clustering-based approach. Simultaneously, the consequent part is designed by Q-value based genetic reinforcement learning. In CQGAF, evolution is performed immediately after the end of one trial in contrast to general GA where many trials are performed before evolution. The feasibility of CQGAF is demonstrated through simulations in cart-pole balancing problems with only binary reinforcement signals.