A Reinforced Hybrid Genetic Algorithm for the Traveling Salesman Problem

We propose a powerful Reinforced Hybrid Genetic Algorithm (RHGA) for the famous NP-hard Traveling Salesman Problem (TSP). RHGA combines reinforcement learning technique with well-known Edge Assembly Crossover genetic algorithm (EAX-GA) and Lin-Kernighan-Helsgaun (LKH) local search heuristic. With help of proposed hybrid mechanism, evolution EAX-GA LKH can boost each other's performance. And based on Q-learning further promotes algorithm. Experimental results 138 widely used TSP benchmarks, number cities ranging from 1,000 to 85,900, demonstrate excellent performance method.