arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.03698cs.AIcs.LGcs.LO
收藏

用于深度强化学习的验证概率策略

Verified Probabilistic Policies for Deep Reinforcement Learning

University of Birmingham(伯明翰大学)

Edoardo Bacci, David Parker

摘要:

深度强化学习是一种越来越流行的技术,用于合成策略以控制代理与其环境的交互。人们对正式验证这些策略是否正确并安全执行的兴趣也越来越大。在这一领域取得的进展是建立在现有工作的基础上,对深层神经网络和连续状态动力系统进行验证。在本文中,我们解决了验证深度强化学习的概率策略的问题,例如,用于解决对抗性环境,打破对称性和管理权衡。我们提出了一个抽象的方法,基于区间马尔可夫决策过程,产生概率保证政策的执行,目前的技术来建立和解决这些模型,使用抽象的解释,混合整数线性规划,基于熵的细化和概率模型检验。我们实现我们的方法,并说明其有效性的选择强化学习基准。

英文摘要:

Deep reinforcement learning is an increasingly popular technique for synthesising policies to control an agent's interaction with its environment. There is also growing interest in formally verifying that such policies are correct and execute safely. Progress has been made in this area by building on existing work for verification of deep neural networks and of continuous-state dynamical systems. In this paper, we tackle the problem of verifying probabilistic policies for deep reinforcement learning, which are used to, for example, tackle adversarial environments, break symmetries and manage trade-offs. We propose an abstraction approach, based on interval Markov decision processes, that yields probabilistic guarantees on a policy's execution, and present techniques to build and solve these models using abstract interpretation, mixed-integer linear programming, entropy-based refinement and probabilistic model checking. We implement our approach and illustrate its effectiveness on a selection of reinforcement learning benchmarks.

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。

↑