A Unified Perspective for Loss-Oriented Imbalanced Learning via Localization
文献类型：期刊论文
| 作者 | Wang, Zitai1; Xu, Qianqian1,8; Yang, Zhiyong9; Xu, Zhikang2,3; Zhang, Linchao4; Cao, Xiaochun5; Huang, Qingming1,6,7 |
| 刊名 | IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
 |
| 出版日期 | 2026 |
| 卷号 | 48期号:1页码:639-656 |
| 关键词 | Calibration Training Safety Predictive models Data mining Computers Additives Smoothing methods Risk minimization Protocols Imbalanced learning re-weighting logit adjustment fisher consistency generalization analysis |
| ISSN号 | 0162-8828 |
| DOI | 10.1109/TPAMI.2025.3609440 |
| 英文摘要 | Due to the inherent imbalance in real-world datasets, na & iuml;ve Empirical Risk Minimization (ERM) tends to bias the learning process towards the majority classes, hindering generalization to minority classes. To rebalance the learning process, one straightforward yet effective approach is to modify the loss function via class-dependent terms, such as re-weighting and logit-adjustment. However, existing analysis of these loss-oriented methods remains coarse-grained and fragmented, failing to explain some empirical results. After reviewing prior work, we find that the properties used through their analysis are typically global, i.e., defined over the whole dataset. Hence, these properties fail to effectively capture how class-dependent terms influence the learning process. To bridge this gap, we turn to explore the localized versions of such properties i.e., defined within each class. Specifically, we employ localized calibration to provide consistency validation across a broader range of losses and localized Lipschitz continuity to provide a fine-grained generalization bound. In this way, we reach a unified perspective for improving and adjusting loss-oriented methods. Finally, a principled learning algorithm is developed based on these insights. Empirical results on both traditional ResNets and foundation models validate our theoretical analyses and demonstrate the effectiveness of the proposed method. |
| 资助项目 | National Natural Science Foundation of China[62525212] ; National Natural Science Foundation of China[62236008] ; National Natural Science Foundation of China[62025604] ; National Natural Science Foundation of China[62441232] ; National Natural Science Foundation of China[U21B2038] ; National Natural Science Foundation of China[U23B2051] ; National Natural Science Foundation of China[62441619] ; Youth Innovation Promotion Association CAS ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDB0680201] ; China National Postdoctoral Program for Innovative Talents[BX20240384] ; Beijing Natural Science Foundation[L252144] ; General Program of the Chinese Postdoctoral Science Foundation[2025M771558] ; Fundamental Research Funds for the Central Universities[E4EQ1101] |
| WOS研究方向 | Computer Science ; Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:001630351400004 |
| 出版者 | IEEE COMPUTER SOC |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/42809]  |
| 专题 | 中国科学院计算技术研究所 |
| 通讯作者 | Xu, Qianqian; Huang, Qingming |
| 作者单位 | 1.Chinese Acad Sci, State Key Lab AI Safety, Inst Comp Technol, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Informat Engn, Beijing 100093, Peoples R China 3.Univ Chinese Acad Sci, Sch Cyber Secur, Beijing 100049, Peoples R China 4.China Elect Technol Grp Corp, Artificial Intelligence Inst, Beijing 100041, Peoples R China 5.Sun Yat Sen Univ, Sch Cyber Sci & Technol, Shenzhen Campus, Shenzhen 518107, Peoples R China 6.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 101408, Peoples R China 7.Univ Chinese Acad Sci, Key Lab Big Data Min & Knowledge Management BDKM, Beijing 101408, Peoples R China 8.Peng Cheng Lab, Shenzhen 518055, Peoples R China 9.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100049, Peoples R China |
| 推荐引用方式 GB/T 7714 | Wang, Zitai,Xu, Qianqian,Yang, Zhiyong,et al. A Unified Perspective for Loss-Oriented Imbalanced Learning via Localization[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2026,48(1):639-656. |
| APA | Wang, Zitai.,Xu, Qianqian.,Yang, Zhiyong.,Xu, Zhikang.,Zhang, Linchao.,...&Huang, Qingming.(2026).A Unified Perspective for Loss-Oriented Imbalanced Learning via Localization.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,48(1),639-656. |
| MLA | Wang, Zitai,et al."A Unified Perspective for Loss-Oriented Imbalanced Learning via Localization".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 48.1(2026):639-656. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
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。