机器学习,功能和目标

IF 0.2 4区 哲学 0 PHILOSOPHY
Patrick Butlin
{"title":"机器学习,功能和目标","authors":"Patrick Butlin","doi":"10.52685/cjp.22.66.5","DOIUrl":null,"url":null,"abstract":"Machine learning researchers distinguish between reinforcement learning and supervised learning and refer to reinforcement learning systems as “agents”. This paper vindicates the claim that systems trained by reinforcement learning are agents while those trained by supervised learning are not. Systems of both kinds satisfy Dretske’s criteria for agency, because they both learn to produce outputs selectively in response to inputs. However, reinforcement learning is sensitive to the instrumental value of outputs, giving rise to systems which exploit the effects of outputs on subsequent inputs to achieve good performance over episodes of interaction with their environments. Supervised learning systems, in contrast, merely learn to produce better outputs in response to individual inputs.","PeriodicalId":43218,"journal":{"name":"Croatian Journal of Philosophy","volume":" ","pages":""},"PeriodicalIF":0.2000,"publicationDate":"2022-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Machine Learning, Functions and Goals\",\"authors\":\"Patrick Butlin\",\"doi\":\"10.52685/cjp.22.66.5\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Machine learning researchers distinguish between reinforcement learning and supervised learning and refer to reinforcement learning systems as “agents”. This paper vindicates the claim that systems trained by reinforcement learning are agents while those trained by supervised learning are not. Systems of both kinds satisfy Dretske’s criteria for agency, because they both learn to produce outputs selectively in response to inputs. However, reinforcement learning is sensitive to the instrumental value of outputs, giving rise to systems which exploit the effects of outputs on subsequent inputs to achieve good performance over episodes of interaction with their environments. Supervised learning systems, in contrast, merely learn to produce better outputs in response to individual inputs.\",\"PeriodicalId\":43218,\"journal\":{\"name\":\"Croatian Journal of Philosophy\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":0.2000,\"publicationDate\":\"2022-12-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Croatian Journal of Philosophy\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.52685/cjp.22.66.5\",\"RegionNum\":4,\"RegionCategory\":\"哲学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"0\",\"JCRName\":\"PHILOSOPHY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Croatian Journal of Philosophy","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.52685/cjp.22.66.5","RegionNum":4,"RegionCategory":"哲学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"0","JCRName":"PHILOSOPHY","Score":null,"Total":0}
引用次数: 2

摘要

机器学习研究人员区分强化学习和监督学习,并将强化学习系统称为“代理”。本文证明了通过强化学习训练的系统是代理,而通过监督学习训练的不是代理的说法。这两种系统都满足Dretske的代理标准,因为它们都学会了根据输入有选择地产生输出。然而,强化学习对输出的工具价值很敏感,导致系统利用输出对后续输入的影响,在与环境的互动中取得良好的表现。相比之下,有监督的学习系统只是学会根据个人输入产生更好的输出。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Machine Learning, Functions and Goals
Machine learning researchers distinguish between reinforcement learning and supervised learning and refer to reinforcement learning systems as “agents”. This paper vindicates the claim that systems trained by reinforcement learning are agents while those trained by supervised learning are not. Systems of both kinds satisfy Dretske’s criteria for agency, because they both learn to produce outputs selectively in response to inputs. However, reinforcement learning is sensitive to the instrumental value of outputs, giving rise to systems which exploit the effects of outputs on subsequent inputs to achieve good performance over episodes of interaction with their environments. Supervised learning systems, in contrast, merely learn to produce better outputs in response to individual inputs.
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
CiteScore
0.20
自引率
0.00%
发文量
15
×
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