网络瓶颈和任务结构控制着觅食机器人可解释学习规则的演化

Emmanouil Giannakakis, Sina Khajehabdollahi, Anna Levina
{"title":"网络瓶颈和任务结构控制着觅食机器人可解释学习规则的演化","authors":"Emmanouil Giannakakis, Sina Khajehabdollahi, Anna Levina","doi":"arxiv-2403.13649","DOIUrl":null,"url":null,"abstract":"Developing reliable mechanisms for continuous local learning is a central\nchallenge faced by biological and artificial systems. Yet, how the\nenvironmental factors and structural constraints on the learning network\ninfluence the optimal plasticity mechanisms remains obscure even for simple\nsettings. To elucidate these dependencies, we study meta-learning via\nevolutionary optimization of simple reward-modulated plasticity rules in\nembodied agents solving a foraging task. We show that unconstrained\nmeta-learning leads to the emergence of diverse plasticity rules. However,\nregularization and bottlenecks to the model help reduce this variability,\nresulting in interpretable rules. Our findings indicate that the meta-learning\nof plasticity rules is very sensitive to various parameters, with this\nsensitivity possibly reflected in the learning rules found in biological\nnetworks. When included in models, these dependencies can be used to discover\npotential objective functions and details of biological learning via\ncomparisons with experimental observations.","PeriodicalId":501305,"journal":{"name":"arXiv - PHYS - Adaptation and Self-Organizing Systems","volume":"102 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Network bottlenecks and task structure control the evolution of interpretable learning rules in a foraging agent\",\"authors\":\"Emmanouil Giannakakis, Sina Khajehabdollahi, Anna Levina\",\"doi\":\"arxiv-2403.13649\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Developing reliable mechanisms for continuous local learning is a central\\nchallenge faced by biological and artificial systems. Yet, how the\\nenvironmental factors and structural constraints on the learning network\\ninfluence the optimal plasticity mechanisms remains obscure even for simple\\nsettings. To elucidate these dependencies, we study meta-learning via\\nevolutionary optimization of simple reward-modulated plasticity rules in\\nembodied agents solving a foraging task. We show that unconstrained\\nmeta-learning leads to the emergence of diverse plasticity rules. However,\\nregularization and bottlenecks to the model help reduce this variability,\\nresulting in interpretable rules. Our findings indicate that the meta-learning\\nof plasticity rules is very sensitive to various parameters, with this\\nsensitivity possibly reflected in the learning rules found in biological\\nnetworks. When included in models, these dependencies can be used to discover\\npotential objective functions and details of biological learning via\\ncomparisons with experimental observations.\",\"PeriodicalId\":501305,\"journal\":{\"name\":\"arXiv - PHYS - Adaptation and Self-Organizing Systems\",\"volume\":\"102 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-03-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Adaptation and Self-Organizing Systems\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2403.13649\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Adaptation and Self-Organizing Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2403.13649","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

开发可靠的持续局部学习机制是生物和人工系统面临的核心挑战。然而,环境因素和学习网络的结构限制如何影响最佳可塑性机制,即使是在简单的环境中,也仍然是模糊不清的。为了阐明这些依赖关系,我们研究了在解决觅食任务的本体中通过进化优化简单奖励调制可塑性规则的元学习。我们的研究表明,无约束元学习会导致多种可塑性规则的出现。然而,模型的正则化和瓶颈有助于减少这种可变性,从而产生可解释的规则。我们的研究结果表明,可塑性规则的元学习对各种参数非常敏感,这种敏感性可能反映在生物网络的学习规则中。如果将这些依赖关系纳入模型,就能发现潜在的目标函数以及生物学习与实验观察比较的细节。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Network bottlenecks and task structure control the evolution of interpretable learning rules in a foraging agent
Developing reliable mechanisms for continuous local learning is a central challenge faced by biological and artificial systems. Yet, how the environmental factors and structural constraints on the learning network influence the optimal plasticity mechanisms remains obscure even for simple settings. To elucidate these dependencies, we study meta-learning via evolutionary optimization of simple reward-modulated plasticity rules in embodied agents solving a foraging task. We show that unconstrained meta-learning leads to the emergence of diverse plasticity rules. However, regularization and bottlenecks to the model help reduce this variability, resulting in interpretable rules. Our findings indicate that the meta-learning of plasticity rules is very sensitive to various parameters, with this sensitivity possibly reflected in the learning rules found in biological networks. When included in models, these dependencies can be used to discover potential objective functions and details of biological learning via comparisons with experimental observations.
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
自引率
0.00%
发文量
0
期刊最新文献
Expected and unexpected routes to synchronization in a system of swarmalators Synchronization cluster bursting in adaptive oscillators networks The forced one-dimensional swarmalator model Periodic systems have new classes of synchronization stability Reduced-order adaptive synchronization in a chaotic neural network with parameter mismatch: A dynamical system vs. machine learning approach
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1