Reinforcement structure/parameter learning for neural-network-based fuzzy logic control systems | Synapse