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 1901.07186cs.LGcs.ROstat.ML
收藏

迈向从单个视频演示中学习模仿

Towards Learning to Imitate from a Single Video Demonstration

Université de Montréal(蒙特利尔大学);Mila Quebec AI Institute(米拉魁北克人工智能研究所);Polytechnique Montréal(蒙特利尔理工学院);ServiceNow Research(ServiceNow研究院)

Glen Berseth, Florian Golemo, Christopher Pal

AI总结:

针对仅凭视频观察且无状态动作信息的模仿学习问题,提出基于 Siamese 循环网络的对比奖励学习方法,并用多任务数据和图像编码损失提升效果,在多种二维、三维模拟运动体上超过现有方法。

摘要:

能够仅凭给定视频观察进行学习模仿——而无法直接获取状态或动作信息——的智能体,更适用于在自然世界中学习。然而,构建一个有助于实现这一目标的强化学习(RL)智能体仍是一项重大挑战。我们通过对比训练来应对这一挑战,以学习一个奖励函数,将智能体的行为与单个演示进行比较。我们使用 Siamese 循环神经网络架构来学习运动片段之间在空间和时间上的奖励,同时训练一个 RL 策略以最小化这一距离。通过实验,我们还发现,纳入多任务数据和额外的图像编码损失能够改善所学奖励的时间一致性,从而显著提升策略学习效果。我们在二维环境中的模拟类人、狗和猛禽智能体,以及三维环境中的四足动物和类人智能体上展示了我们的方法。结果表明,我们的方法在这些环境中优于当前最先进的技术,并且能够从单个视频演示中学习模仿。

英文摘要:

Agents that can learn to imitate given video observation -- \emph{without direct access to state or action information} are more applicable to learning in the natural world. However, formulating a reinforcement learning (RL) agent that facilitates this goal remains a significant challenge. We approach this challenge using contrastive training to learn a reward function comparing an agent's behaviour with a single demonstration. We use a Siamese recurrent neural network architecture to learn rewards in space and time between motion clips while training an RL policy to minimize this distance. Through experimentation, we also find that the inclusion of multi-task data and additional image encoding losses improve the temporal consistency of the learned rewards and, as a result, significantly improves policy learning. We demonstrate our approach on simulated humanoid, dog, and raptor agents in 2D and a quadruped and a humanoid in 3D. We show that our method outperforms current state-of-the-art techniques in these environments and can learn to imitate from a single video demonstration.

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。

↑