2019/05/31 by Wen-Ji Zhou, Yang Yu, Zhou, Wen-Ji +12
Computer Science · Engineering · Mathematics · #Artificial Intelligence in Games #Artificial intelligence #Computer science #Computer security #Engineering #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Granularity #Grid #Human–computer interaction #Key (lock) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine learning #Reinforcement Learning in Robotics #Reinforcement learning #Representation (politics) #Residual #Reuse #Set (abstract data type) #Task (project management) #cs.LG #stat.ML
paper · pdf · doi:10.48550/arxiv.1905.13719
Conference version appears in IJCAI 2019
arxiv created 2019/05/31 · openalex publication_date 2019/05/31 · arxiv updated 2019/06/03 · openalex created_date 2019/06/07 · openalex updated_date 2026/07/28
Experience reuse is key to sample-efficient reinforcement learning. One of the critical issues is how the experience is represented and stored. Previously, the experience can be stored in the forms of features, individual models, and the average model, each lying at a different granularity. However, new tasks may require experience across multiple granularities. In this paper, we propose the policy residual representation (PRR) network, which can extract and store multiple levels of experience. PRR network is trained on a set of tasks with a multi-level architecture, where a module in each level corresponds to a subset of the tasks. Therefore, the PRR network represents the experience in a spectrum-like way. When training on a new task, PRR can provide different levels of experience for accelerating the learning. We experiment with the PRR network on a set of grid world navigation tasks, locomotion tasks, and fighting tasks in a video game. The results show that the PRR network leads to better reuse of experience and thus outperforms some state-of-the-art approaches.