arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.03006cs.LG
收藏

强化学习快速生成模型的学习与查询

Learning and Querying Fast Generative Models for Reinforcement Learning

DeepMind

Lars Buesing, Theophane Weber, Sebastien Racaniere, S. M. Ali Eslami, Danilo Rezende, David P. Reichert, Fabio Viola, Frederic Besse, Karol Gregor, Demis Hassabis, Daan Wierstra

摘要:

基于基于模型的强化学习(RL)的一个关键挑战是综合计算效率高且准确的环境模型。我们表明,精心设计的生成模型,学习和操作紧凑的状态表示,所谓的状态空间模型,大大降低了预测结果的行动序列的计算成本。大量的实验表明,状态空间模型准确地捕捉动态的雅达利游戏从街机学习环境从原始像素。状态空间模型的计算速度提高,同时保持高精度,使其在强化学习中的应用是可行的:我们证明,代理查询这些模型的决策优于强大的无模型基线的游戏MSPACMAN,展示了使用学习环境模型进行规划的潜力。

英文摘要:

A key challenge in model-based reinforcement learning (RL) is to synthesize computationally efficient and accurate environment models. We show that carefully designed generative models that learn and operate on compact state representations, so-called state-space models, substantially reduce the computational costs for predicting outcomes of sequences of actions. Extensive experiments establish that state-space models accurately capture the dynamics of Atari games from the Arcade Learning Environment from raw pixels. The computational speed-up of state-space models while maintaining high accuracy makes their application in RL feasible: We demonstrate that agents which query these models for decision making outperform strong model-free baselines on the game MSPACMAN, demonstrating the potential of using learned environment models for planning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑