arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.09337cs.LGcs.AI
收藏

几个后向模型演示的鲁棒模拟

Robust Imitation of a Few Demonstrations with a Backwards Model

Khoury College of Computer Sciences(库里计算机学院);Northeastern University(东北大学)

Jung Yeon Park, Lawson L. S. Wong

摘要:

与强化学习相比,专家演示的行为克隆可以以更有效的样本方式加速学习最优策略。然而,该策略不能很好地外推到演示数据之外的看不见的状态,从而产生协变量偏移(代理偏离演示)和复合误差。在这项工作中,我们通过扩展演示周围的吸引区域来解决这个问题,以便代理可以学习如何在偏离路线时回到演示的轨迹上。我们训练一个生成向后动力学模型,并从演示中的状态生成短的想象轨迹。通过模仿演示和这些模型的展示,智能体学习演示的路径以及如何回到这些路径上。通过最优或接近最优的演示,学习到的策略将是最优的,并且对偏差具有鲁棒性,具有更宽的吸引区域。在连续控制域上,我们评估了从演示数据中看不到的不同初始状态开始时的鲁棒性。虽然我们的方法和其他模仿学习基线都可以成功地解决训练分布中初始状态的任务,但我们的方法对不同的初始状态表现出更强的鲁棒性。

英文摘要:

Behavior cloning of expert demonstrations can speed up learning optimal policies in a more sample-efficient way over reinforcement learning. However, the policy cannot extrapolate well to unseen states outside of the demonstration data, creating covariate shift (agent drifting away from demonstrations) and compounding errors. In this work, we tackle this issue by extending the region of attraction around the demonstrations so that the agent can learn how to get back onto the demonstrated trajectories if it veers off-course. We train a generative backwards dynamics model and generate short imagined trajectories from states in the demonstrations. By imitating both demonstrations and these model rollouts, the agent learns the demonstrated paths and how to get back onto these paths. With optimal or near-optimal demonstrations, the learned policy will be both optimal and robust to deviations, with a wider region of attraction. On continuous control domains, we evaluate the robustness when starting from different initial states unseen in the demonstration data. While both our method and other imitation learning baselines can successfully solve the tasks for initial states in the training distribution, our method exhibits considerably more robustness to different initial states.

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。

↑