When Collaborative Filtering Meets Reinforcement Learning

Yu Lei, Wenjie Li

In this paper, we study a multi-step interactive recommendation problem, where the item recommended at current step may affect the quality of future recommendations. To address the…