A Reinforcement Learning based Collection Approach | IEEE Conference Publication | IEEE Xplore