NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:8751
Title:Search on the Replay Buffer: Bridging Planning and Reinforcement Learning

The paper presents a general-purpose control algorithm combining planning and RL to solve tasks with sparse rewards or with long horizon. This algorithm is novel and interesting. The three reviewers agree that the contributions presented here should be published at the conference. The rebuttal helped solving most clarification issues. The reviewers also suggest various ways to further improve the manuscript. These include: - A more detailed discussion on the types of tasks the method could efficiently solve. - A discussion on how the replay buffer could be designed and optimized. - A more precise description of the algorithm and of the experiments.