arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1709.07643cs.ROcs.AI
收藏

OptLayer -现实世界中深度强化学习的实用约束优化

OptLayer - Practical Constrained Optimization for Deep Reinforcement Learning in the Real World

IBM Research AI(IBM研究院人工智能部门)

Tu-Hoa Pham, Giovanni De Magistris, Ryuki Tachibana

摘要:

虽然深度强化学习技术最近在许多决策问题上取得了相当大的成就,但它们在机器人技术中的使用在很大程度上仅限于模拟世界或受限运动,因为现实世界中不受约束的试错交互可能对机器人或其环境产生不良后果。为了克服这些限制,我们提出了一种新的强化学习架构OptLayer,它将神经网络预测的可能不安全的动作作为输入,并输出满足所选约束的最接近的动作。虽然学习控制策略通常需要仔细制定奖励和惩罚,同时探索可能的操作范围,但OptLayer确保在训练期间只执行安全的操作,并惩罚不安全的预测。我们证明了我们的方法对机器人达到任务的有效性,无论是模拟还是在现实世界中。

英文摘要:

While deep reinforcement learning techniques have recently produced considerable achievements on many decision-making problems, their use in robotics has largely been limited to simulated worlds or restricted motions, since unconstrained trial-and-error interactions in the real world can have undesirable consequences for the robot or its environment. To overcome such limitations, we propose a novel reinforcement learning architecture, OptLayer, that takes as inputs possibly unsafe actions predicted by a neural network and outputs the closest actions that satisfy chosen constraints. While learning control policies often requires carefully crafted rewards and penalties while exploring the range of possible actions, OptLayer ensures that only safe actions are actually executed and unsafe predictions are penalized during training. We demonstrate the effectiveness of our approach on robot reaching tasks, both simulated and in the real world.

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。

↑