RLBS: An Adaptive Backtracking Strategy Based on Reinforcement Learning for Combinatorial Optimization | IEEE Conference Publication | IEEE Xplore