arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1707.02201cs.ROcs.LGcs.SYeess.SY
收藏

通过对抗模仿学习学习人类行为

Learning human behaviors from motion capture by adversarial imitation

DeepMind(深度思维)

Josh Merel, Yuval Tassa, Dhruva TB, Sriram Srinivasan, Jay Lemmon, Ziyu Wang, Greg Wayne, Nicolas Heess

AI总结:

本文提出利用生成对抗模仿学习训练神经网络策略,从有限的不完全观测状态特征中生成人类化运动模式,即使演示来自不同物理参数的躯体,也能通过子技能策略解决任务。

摘要:

深度强化学习的快速进展使训练高维人形身体控制器变得越来越可行。然而,纯强化学习方法使用简单的奖励函数往往会产生非人类化且过于刻板的运动行为。在本文中,我们扩展了生成对抗模仿学习,以使训练通用神经网络策略成为可能,从而从仅包含部分观测状态特征的有限演示中生成人类化运动模式,即使在没有动作信息且演示来自具有不同且未知物理参数的躯体时也是如此。我们利用这种方法从动作捕捉数据构建子技能策略,并展示这些策略在由更高层次控制器控制时可以用于解决任务。

英文摘要:

Rapid progress in deep reinforcement learning has made it increasingly feasible to train controllers for high-dimensional humanoid bodies. However, methods that use pure reinforcement learning with simple reward functions tend to produce non-humanlike and overly stereotyped movement behaviors. In this work, we extend generative adversarial imitation learning to enable training of generic neural network policies to produce humanlike movement patterns from limited demonstrations consisting only of partially observed state features, without access to actions, even when the demonstrations come from a body with different and unknown physical parameters. We leverage this approach to build sub-skill policies from motion capture data and show that they can be reused to solve tasks when controlled by a higher level controller.

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。

↑