Learning State-Specific Action Masks for Reinforcement Learning | |
Wang ZY(王梓薏)4,5; Li XR(李欣然)4,5; Sun LY(孙罗洋)4,5; Zhang HF(张海峰)3,4,5; Liu HL(刘华林)2; Jun Wang1 | |
刊名 | Algorithms |
2024-01 | |
卷号 | 17期号:2页码:60 |
关键词 | reinforcement learning exploration efficiency space reduction |
英文摘要 | Efficient yet sufficient exploration remains a critical challenge in reinforcement learning (RL), especially for Markov Decision Processes (MDPs) with vast action spaces. Previous approaches have commonly involved projecting the original action space into a latent space or employing environmental action masks to reduce the action possibilities. Nevertheless, these methods often lack interpretability or rely on expert knowledge. In this study, we introduce a novel method for automatically reducing the action space in environments with discrete action spaces while preserving interpretability. The proposed approach learns state-specific masks with a dual purpose: (1) eliminating actions with minimal influence on the MDP and (2) aggregating actions with identical behavioral consequences within the MDP. Specifically, we introduce a novel concept called Bisimulation Metrics on Actions by States (BMAS) to quantify the behavioral consequences of actions within the MDP and design a dedicated mask model to ensure their binary nature. Crucially, we present a practical learning procedure for training the mask model, leveraging transition data collected by any RL policy. Our method is designed to be plug-and-play and adaptable to all RL policies, and to validate its effectiveness, an integration into two prominent RL algorithms, DQN and PPO, is performed. Experimental results obtained from Maze, Atari, and RTS2 reveal a substantial acceleration in the RL learning process and noteworthy performance improvements facilitated by the introduced approach. |
URL标识 | 查看原文 |
内容类型 | 期刊论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/58507] |
专题 | 复杂系统认知与决策实验室_群体决策智能团队 |
通讯作者 | Zhang HF(张海峰) |
作者单位 | 1.Computer Science, University College London, London WC1E 6BT, UK 2.Key Laboratory of Oil & Gas Business Chain Optimization, Petrochina Planning and Engineering Institute, CNPC, Beijing 100083, China 3.Nanjing Artificial Intelligence Research of IA, Jiangning District, Nanjing 211135, China 4.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 101408, China 5.Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China |
推荐引用方式 GB/T 7714 | Wang ZY,Li XR,Sun LY,et al. Learning State-Specific Action Masks for Reinforcement Learning[J]. Algorithms,2024,17(2):60. |
APA | Wang ZY,Li XR,Sun LY,Zhang HF,Liu HL,&Jun Wang.(2024).Learning State-Specific Action Masks for Reinforcement Learning.Algorithms,17(2),60. |
MLA | Wang ZY,et al."Learning State-Specific Action Masks for Reinforcement Learning".Algorithms 17.2(2024):60. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论