泰勒共识网络:拥抱噪音,揭示科学数据的不确定性

Guangxuan Song, Dongmei Fu, Zhongwei Qiu, Jintao Meng, Dawei Zhang
{"title":"泰勒共识网络:拥抱噪音,揭示科学数据的不确定性","authors":"Guangxuan Song, Dongmei Fu, Zhongwei Qiu, Jintao Meng, Dawei Zhang","doi":"arxiv-2409.07942","DOIUrl":null,"url":null,"abstract":"Uncertainty estimation is crucial in scientific data for machine learning.\nCurrent uncertainty estimation methods mainly focus on the model's inherent\nuncertainty, while neglecting the explicit modeling of noise in the data.\nFurthermore, noise estimation methods typically rely on temporal or spatial\ndependencies, which can pose a significant challenge in structured scientific\ndata where such dependencies among samples are often absent. To address these\nchallenges in scientific research, we propose the Taylor-Sensus Network\n(TSNet). TSNet innovatively uses a Taylor series expansion to model complex,\nheteroscedastic noise and proposes a deep Taylor block for aware noise\ndistribution. TSNet includes a noise-aware contrastive learning module and a\ndata density perception module for aleatoric and epistemic uncertainty.\nAdditionally, an uncertainty combination operator is used to integrate these\nuncertainties, and the network is trained using a novel heteroscedastic mean\nsquare error loss. TSNet demonstrates superior performance over mainstream and\nstate-of-the-art methods in experiments, highlighting its potential in\nscientific research and noise resistance. It will be open-source to facilitate\nthe community of \"AI for Science\".","PeriodicalId":501301,"journal":{"name":"arXiv - CS - Machine Learning","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Taylor-Sensus Network: Embracing Noise to Enlighten Uncertainty for Scientific Data\",\"authors\":\"Guangxuan Song, Dongmei Fu, Zhongwei Qiu, Jintao Meng, Dawei Zhang\",\"doi\":\"arxiv-2409.07942\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Uncertainty estimation is crucial in scientific data for machine learning.\\nCurrent uncertainty estimation methods mainly focus on the model's inherent\\nuncertainty, while neglecting the explicit modeling of noise in the data.\\nFurthermore, noise estimation methods typically rely on temporal or spatial\\ndependencies, which can pose a significant challenge in structured scientific\\ndata where such dependencies among samples are often absent. To address these\\nchallenges in scientific research, we propose the Taylor-Sensus Network\\n(TSNet). TSNet innovatively uses a Taylor series expansion to model complex,\\nheteroscedastic noise and proposes a deep Taylor block for aware noise\\ndistribution. TSNet includes a noise-aware contrastive learning module and a\\ndata density perception module for aleatoric and epistemic uncertainty.\\nAdditionally, an uncertainty combination operator is used to integrate these\\nuncertainties, and the network is trained using a novel heteroscedastic mean\\nsquare error loss. TSNet demonstrates superior performance over mainstream and\\nstate-of-the-art methods in experiments, highlighting its potential in\\nscientific research and noise resistance. It will be open-source to facilitate\\nthe community of \\\"AI for Science\\\".\",\"PeriodicalId\":501301,\"journal\":{\"name\":\"arXiv - CS - Machine Learning\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Machine Learning\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07942\",\"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 - CS - Machine Learning","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07942","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

目前的不确定性估计方法主要关注模型的固有不确定性,而忽略了数据中噪声的显式建模。此外,噪声估计方法通常依赖于时间或空间依赖性,这对于结构化科学数据来说是一个巨大的挑战,因为样本之间往往不存在这种依赖性。为了解决科学研究中的这些挑战,我们提出了泰勒-共识网络(TSNet)。TSNet 创新性地使用泰勒级数展开来模拟复杂的异速噪声,并为感知噪声分布提出了深度泰勒块。TSNet 包括一个噪声感知对比学习模块和一个数据密度感知模块,用于分析不确定性和认识不确定性。此外,还使用了一个不确定性组合算子来整合这些不确定性,并使用一种新颖的异前缀均方误差损失来训练网络。在实验中,TSNet 的性能优于主流方法和最先进的方法,凸显了其在科学研究和抗噪声方面的潜力。它将开源,以促进 "人工智能促进科学 "社区的发展。
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
Taylor-Sensus Network: Embracing Noise to Enlighten Uncertainty for Scientific Data
Uncertainty estimation is crucial in scientific data for machine learning. Current uncertainty estimation methods mainly focus on the model's inherent uncertainty, while neglecting the explicit modeling of noise in the data. Furthermore, noise estimation methods typically rely on temporal or spatial dependencies, which can pose a significant challenge in structured scientific data where such dependencies among samples are often absent. To address these challenges in scientific research, we propose the Taylor-Sensus Network (TSNet). TSNet innovatively uses a Taylor series expansion to model complex, heteroscedastic noise and proposes a deep Taylor block for aware noise distribution. TSNet includes a noise-aware contrastive learning module and a data density perception module for aleatoric and epistemic uncertainty. Additionally, an uncertainty combination operator is used to integrate these uncertainties, and the network is trained using a novel heteroscedastic mean square error loss. TSNet demonstrates superior performance over mainstream and state-of-the-art methods in experiments, highlighting its potential in scientific research and noise resistance. It will be open-source to facilitate the community of "AI for Science".
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
Almost Sure Convergence of Linear Temporal Difference Learning with Arbitrary Features The Impact of Element Ordering on LM Agent Performance Towards Interpretable End-Stage Renal Disease (ESRD) Prediction: Utilizing Administrative Claims Data with Explainable AI Techniques Extended Deep Submodular Functions Symmetry-Enriched Learning: A Category-Theoretic Framework for Robust Machine Learning Models
×
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