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 2310.09574cs.LG
收藏

硬约束连续控制的约简策略优化

Reduced Policy Optimization for Continuous Control with Hard Constraints

ShanghaiTech University(上海科技大学);King’s College London(伦敦国王学院)

Shutong Ding, Jingya Wang, Yali Du, Ye Shi

AI总结:

针对一般硬约束(含非凸)的连续控制任务,提出结合广义约简梯度与强化学习的约简策略优化算法,通过划分基本/非基本动作求解约束,并引入投影与拉格朗日松弛,在三个新基准上优于现有方法。

摘要:

约束强化学习(RL)的最新进展赋予了强化学习一定的安全保证。然而,在具有一般硬约束的连续控制任务中部署现有的约束强化学习算法仍然具有挑战性,尤其是在那些具有非凸硬约束的情况下。受经典约束优化技术——广义约简梯度(GRG)算法的启发,我们提出了一种约简策略优化(RPO)算法,该算法将强化学习与GRG相结合以处理一般硬约束。RPO按照GRG方法将动作划分为基本动作和非基本动作,并通过策略网络输出基本动作。随后,RPO利用获得的基本动作,通过求解基于等式约束的方程来计算非基本动作。然后,通过隐式微分非基本动作关于基本动作来更新策略网络。此外,我们引入了一种基于约简梯度的动作投影过程,并应用改进的拉格朗日松弛技术来确保不等式约束得到满足。据我们所知,RPO是首次将GRG引入强化学习,作为一种有效处理等式和不等式硬约束的方法。值得注意的是,目前缺乏具有复杂硬约束的强化学习环境,这促使我们开发了三个新的基准:两个机器人操作任务和一个智能电网运行控制任务。利用这些基准,RPO在累积奖励和约束违反方面均优于先前的约束强化学习算法。我们相信,RPO连同新的基准,将为将强化学习应用于具有复杂约束的现实世界问题开辟新的机遇。

英文摘要:

Recent advances in constrained reinforcement learning (RL) have endowed reinforcement learning with certain safety guarantees. However, deploying existing constrained RL algorithms in continuous control tasks with general hard constraints remains challenging, particularly in those situations with non-convex hard constraints. Inspired by the generalized reduced gradient (GRG) algorithm, a classical constrained optimization technique, we propose a reduced policy optimization (RPO) algorithm that combines RL with GRG to address general hard constraints. RPO partitions actions into basic actions and nonbasic actions following the GRG method and outputs the basic actions via a policy network. Subsequently, RPO calculates the nonbasic actions by solving equations based on equality constraints using the obtained basic actions. The policy network is then updated by implicitly differentiating nonbasic actions with respect to basic actions. Additionally, we introduce an action projection procedure based on the reduced gradient and apply a modified Lagrangian relaxation technique to ensure inequality constraints are satisfied. To the best of our knowledge, RPO is the first attempt that introduces GRG to RL as a way of efficiently handling both equality and inequality hard constraints. It is worth noting that there is currently a lack of RL environments with complex hard constraints, which motivates us to develop three new benchmarks: two robotics manipulation tasks and a smart grid operation control task. With these benchmarks, RPO achieves better performance than previous constrained RL algorithms in terms of both cumulative reward and constraint violation. We believe RPO, along with the new benchmarks, will open up new opportunities for applying RL to real-world problems with complex constraints.

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。

↑