arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2008.10066cs.LGcs.AIcs.RO
收藏

通过在线规划学习非策略

Learning Off-Policy with Online Planning

Robotics Institute(机器人研究所);Carnegie Mellon University(卡内基梅隆大学);The University of Texas at Austin(得克萨斯大学奥斯汀分校)

Harshit Sikchi, Wenxuan Zhou, David Held

摘要:

低数据和风险敏感域中的强化学习(RL)需要高性能和灵活的部署策略,这些策略可以在部署过程中轻松纳入约束。一类这样的策略是半参数H步前瞻策略,其在具有终端价值函数的固定时间范围的动态模型上使用轨迹优化来选择动作。在这项工作中,我们研究了一种新的实例化的H-步骤前瞻与学习模型和终端价值函数学习的无模型的离线策略算法,名为学习离线策略与在线规划(LOOP)。我们对这种方法进行了理论分析,提出了模型误差和价值函数误差之间的权衡,并实证证明了这种权衡在深度强化学习中是有益的。此外,我们确定在这个框架中的“演员发散”的问题,并提出演员正则化控制(ARC),修改后的轨迹优化过程。我们评估了我们的方法对一组机器人任务离线和在线RL,并展示了改进的性能。我们还展示了LOOP的灵活性,在部署过程中将安全约束与一组导航环境。我们证明,循环是一个理想的框架,机器人应用程序的基础上,其强大的性能在各种重要的RL设置。项目视频和详细信息可以在https://hari-sikchi.github.io/loop上找到。

英文摘要:

Reinforcement learning (RL) in low-data and risk-sensitive domains requires performant and flexible deployment policies that can readily incorporate constraints during deployment. One such class of policies are the semi-parametric H-step lookahead policies, which select actions using trajectory optimization over a dynamics model for a fixed horizon with a terminal value function. In this work, we investigate a novel instantiation of H-step lookahead with a learned model and a terminal value function learned by a model-free off-policy algorithm, named Learning Off-Policy with Online Planning (LOOP). We provide a theoretical analysis of this method, suggesting a tradeoff between model errors and value function errors and empirically demonstrate this tradeoff to be beneficial in deep reinforcement learning. Furthermore, we identify the "Actor Divergence" issue in this framework and propose Actor Regularized Control (ARC), a modified trajectory optimization procedure. We evaluate our method on a set of robotic tasks for Offline and Online RL and demonstrate improved performance. We also show the flexibility of LOOP to incorporate safety constraints during deployment with a set of navigation environments. We demonstrate that LOOP is a desirable framework for robotics applications based on its strong performance in various important RL settings. Project video and details can be found at https://hari-sikchi.github.io/loop .

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。

↑