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 2003.09398cs.LGcs.ROstat.ML
收藏

深度约束Q学习

Deep Constrained Q-learning

University of Freiburg(弗赖堡大学);BrainLinks-BrainTools(脑连接-脑工具)

Gabriel Kalweit, Maria Huegle, Moritz Werling, Joschka Boedecker

摘要:

在许多现实世界的应用中,强化学习代理必须优化多个目标,同时遵循某些规则或满足一系列约束。基于奖励塑造的经典方法,即奖励信号中不同目标的加权组合,或拉格朗日方法,包括损失函数中的约束,不能保证代理在所有时间点都满足约束,并且可能导致不期望的行为。当从动作-价值函数中提取离散策略时,可以通过将动作空间限制在最大化来确保安全动作,但是可能导致可行替代方案中的次优解。在这项工作中,我们提出了约束Q-学习,一种新的政策外强化学习框架,直接在Q-更新中限制动作空间,以学习诱导的约束MDP和相应的安全策略的最佳Q-函数。除了单步约束只涉及到下一个动作,我们介绍了一个制定近似的多步约束下的当前目标政策的基础上截断值函数。我们分析了约束Q学习在表格情况下的优势,并将约束DQN与奖励塑造和拉格朗日方法在自动驾驶高级决策中的应用进行了比较,考虑了安全性,保持正确性和舒适性的约束。我们在开源模拟器SUMO和真实的HighD数据集上训练我们的代理。

英文摘要:

In many real world applications, reinforcement learning agents have to optimize multiple objectives while following certain rules or satisfying a list of constraints. Classical methods based on reward shaping, i.e. a weighted combination of different objectives in the reward signal, or Lagrangian methods, including constraints in the loss function, have no guarantees that the agent satisfies the constraints at all points in time and can lead to undesired behavior. When a discrete policy is extracted from an action-value function, safe actions can be ensured by restricting the action space at maximization, but can lead to sub-optimal solutions among feasible alternatives. In this work, we propose Constrained Q-learning, a novel off-policy reinforcement learning framework restricting the action space directly in the Q-update to learn the optimal Q-function for the induced constrained MDP and the corresponding safe policy. In addition to single-step constraints referring only to the next action, we introduce a formulation for approximate multi-step constraints under the current target policy based on truncated value-functions. We analyze the advantages of Constrained Q-learning in the tabular case and compare Constrained DQN to reward shaping and Lagrangian methods in the application of high-level decision making in autonomous driving, considering constraints for safety, keeping right and comfort. We train our agent in the open-source simulator SUMO and on the real HighD data set.

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。

↑