Learning to Optimize Vehicle Routes Problem: A Two-Stage Hybrid Reinforcement Learning | IEEE Conference Publication | IEEE Xplore