深度强化学习中的符号任务推理

IF 4.5 3区 计算机科学 Q2 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Hosein Hasanbeig, N. Jeppu, Alessandro Abate, Tom Melham, Daniel Kroening
{"title":"深度强化学习中的符号任务推理","authors":"Hosein Hasanbeig, N. Jeppu, Alessandro Abate, Tom Melham, Daniel Kroening","doi":"10.1613/jair.1.14063","DOIUrl":null,"url":null,"abstract":"This paper proposes DeepSynth, a method for effective training of deep reinforcement learning agents when the reward is sparse or non-Markovian, but at the same time progress towards the reward requires achieving an unknown sequence of high-level objectives. Our method employs a novel algorithm for synthesis of compact finite state automata to uncover this sequential structure automatically. We synthesise a human-interpretable automaton from trace data collected by exploring the environment. The state space of the environment is then enriched with the synthesised automaton, so that the generation of a control policy by deep reinforcement learning is guided by the discovered structure encoded in the automaton. The proposed approach is able to cope with both high-dimensional, low-level features and unknown sparse or non-Markovian rewards. We have evaluated DeepSynth’s performance in a set of experiments that includes the Atari game Montezuma’s Revenge, known to be challenging. Compared to approaches that rely solely on deep reinforcement learning, we obtain a reduction of two orders of magnitude in the iterations required for policy synthesis, and a significant improvement in scalability.","PeriodicalId":54877,"journal":{"name":"Journal of Artificial Intelligence Research","volume":null,"pages":null},"PeriodicalIF":4.5000,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Symbolic Task Inference in Deep Reinforcement Learning\",\"authors\":\"Hosein Hasanbeig, N. Jeppu, Alessandro Abate, Tom Melham, Daniel Kroening\",\"doi\":\"10.1613/jair.1.14063\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper proposes DeepSynth, a method for effective training of deep reinforcement learning agents when the reward is sparse or non-Markovian, but at the same time progress towards the reward requires achieving an unknown sequence of high-level objectives. Our method employs a novel algorithm for synthesis of compact finite state automata to uncover this sequential structure automatically. We synthesise a human-interpretable automaton from trace data collected by exploring the environment. The state space of the environment is then enriched with the synthesised automaton, so that the generation of a control policy by deep reinforcement learning is guided by the discovered structure encoded in the automaton. The proposed approach is able to cope with both high-dimensional, low-level features and unknown sparse or non-Markovian rewards. We have evaluated DeepSynth’s performance in a set of experiments that includes the Atari game Montezuma’s Revenge, known to be challenging. Compared to approaches that rely solely on deep reinforcement learning, we obtain a reduction of two orders of magnitude in the iterations required for policy synthesis, and a significant improvement in scalability.\",\"PeriodicalId\":54877,\"journal\":{\"name\":\"Journal of Artificial Intelligence Research\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":4.5000,\"publicationDate\":\"2024-07-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Artificial Intelligence Research\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://doi.org/10.1613/jair.1.14063\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Artificial Intelligence Research","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1613/jair.1.14063","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

摘要

本文提出了一种有效训练深度强化学习代理的方法--DeepSynth,当奖励是稀疏的或非马尔可夫的,但同时为获得奖励需要实现一连串未知的高级目标。我们的方法采用了一种用于合成紧凑型有限状态自动机的新型算法,以自动发现这种序列结构。我们从探索环境收集到的轨迹数据中合成一个人类可理解的自动机。然后,用合成的自动机丰富环境的状态空间,这样,通过深度强化学习生成的控制策略就能以自动机中编码的已发现结构为指导。所提出的方法既能处理高维、低级特征,也能处理未知的稀疏或非马尔可夫奖励。我们在一组实验中对 DeepSynth 的性能进行了评估,其中包括雅达利游戏 "蒙特祖玛的复仇"(Montezuma's Revenge)。与完全依赖深度强化学习的方法相比,我们发现策略合成所需的迭代次数减少了两个数量级,可扩展性也有了显著提高。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Symbolic Task Inference in Deep Reinforcement Learning
This paper proposes DeepSynth, a method for effective training of deep reinforcement learning agents when the reward is sparse or non-Markovian, but at the same time progress towards the reward requires achieving an unknown sequence of high-level objectives. Our method employs a novel algorithm for synthesis of compact finite state automata to uncover this sequential structure automatically. We synthesise a human-interpretable automaton from trace data collected by exploring the environment. The state space of the environment is then enriched with the synthesised automaton, so that the generation of a control policy by deep reinforcement learning is guided by the discovered structure encoded in the automaton. The proposed approach is able to cope with both high-dimensional, low-level features and unknown sparse or non-Markovian rewards. We have evaluated DeepSynth’s performance in a set of experiments that includes the Atari game Montezuma’s Revenge, known to be challenging. Compared to approaches that rely solely on deep reinforcement learning, we obtain a reduction of two orders of magnitude in the iterations required for policy synthesis, and a significant improvement in scalability.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Journal of Artificial Intelligence Research
Journal of Artificial Intelligence Research 工程技术-计算机:人工智能
CiteScore
9.60
自引率
4.00%
发文量
98
审稿时长
4 months
期刊介绍: JAIR(ISSN 1076 - 9757) covers all areas of artificial intelligence (AI), publishing refereed research articles, survey articles, and technical notes. Established in 1993 as one of the first electronic scientific journals, JAIR is indexed by INSPEC, Science Citation Index, and MathSciNet. JAIR reviews papers within approximately three months of submission and publishes accepted articles on the internet immediately upon receiving the final versions. JAIR articles are published for free distribution on the internet by the AI Access Foundation, and for purchase in bound volumes by AAAI Press.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信