Abstract: In recent years, reinforcement learning (RL) becomes a promising technique for solving combinatorial optimization (CO) problems. The advantage of RL for solving CO problems is its powerful ...