Progressive Learning Strategy for Few-Shot Class-Incremental Learning

IF 9.4 1区 计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS IEEE Transactions on Cybernetics Pub Date : 2025-01-22 DOI:10.1109/TCYB.2025.3525724
Kai Hu;Yunjiang Wang;Yuan Zhang;Xieping Gao
{"title":"Progressive Learning Strategy for Few-Shot Class-Incremental Learning","authors":"Kai Hu;Yunjiang Wang;Yuan Zhang;Xieping Gao","doi":"10.1109/TCYB.2025.3525724","DOIUrl":null,"url":null,"abstract":"The goal of few-shot class incremental learning (FSCIL) is to learn new concepts from a limited number of novel samples while preserving the knowledge of previously learned classes. The mainstream FSCIL framework begins with training in the base session, after which the feature extractor is frozen to accommodate novel classes. We observed that traditional base-session training approaches often lead to overfitting on challenging samples, which can lead to reduced robustness in the decision boundaries and exacerbate the forgetting phenomenon when introducing incremental data. To address this issue, we proposed the progressive learning strategy (PGLS). First, inspired by curriculum learning, we developed a covariance noise perturbation approach based on the statistical information as a difficulty measure for assessing sample robustness. We then reweighted the samples based on their robustness, initially concentrating on enhancing model stability by prioritizing robust samples and subsequently leveraging weakly robust samples to improve generalization. Second, we predefined forward compatibility for various virtual class augmentation models. Within base class training, we employed a curriculum learning strategy that progressively introduced fewer to more virtual classes in order to mitigate any adverse effects on model performance. This strategy enhances the adaptability of base classes to novel ones and alleviates forgetting problems. Finally, extensive experiments conducted on the CUB200, CIFAR100, and miniImageNet datasets demonstrate the significant advantages of our proposed method over state-of-the-art models.","PeriodicalId":13112,"journal":{"name":"IEEE Transactions on Cybernetics","volume":"55 3","pages":"1210-1223"},"PeriodicalIF":9.4000,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Cybernetics","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10849630/","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}
引用次数: 0

Abstract

The goal of few-shot class incremental learning (FSCIL) is to learn new concepts from a limited number of novel samples while preserving the knowledge of previously learned classes. The mainstream FSCIL framework begins with training in the base session, after which the feature extractor is frozen to accommodate novel classes. We observed that traditional base-session training approaches often lead to overfitting on challenging samples, which can lead to reduced robustness in the decision boundaries and exacerbate the forgetting phenomenon when introducing incremental data. To address this issue, we proposed the progressive learning strategy (PGLS). First, inspired by curriculum learning, we developed a covariance noise perturbation approach based on the statistical information as a difficulty measure for assessing sample robustness. We then reweighted the samples based on their robustness, initially concentrating on enhancing model stability by prioritizing robust samples and subsequently leveraging weakly robust samples to improve generalization. Second, we predefined forward compatibility for various virtual class augmentation models. Within base class training, we employed a curriculum learning strategy that progressively introduced fewer to more virtual classes in order to mitigate any adverse effects on model performance. This strategy enhances the adaptability of base classes to novel ones and alleviates forgetting problems. Finally, extensive experiments conducted on the CUB200, CIFAR100, and miniImageNet datasets demonstrate the significant advantages of our proposed method over state-of-the-art models.
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
渐进式学习策略在少课-增量学习中的应用
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
IEEE Transactions on Cybernetics
IEEE Transactions on Cybernetics COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE-COMPUTER SCIENCE, CYBERNETICS
CiteScore
25.40
自引率
11.00%
发文量
1869
期刊介绍: The scope of the IEEE Transactions on Cybernetics includes computational approaches to the field of cybernetics. Specifically, the transactions welcomes papers on communication and control across machines or machine, human, and organizations. The scope includes such areas as computational intelligence, computer vision, neural networks, genetic algorithms, machine learning, fuzzy systems, cognitive systems, decision making, and robotics, to the extent that they contribute to the theme of cybernetics or demonstrate an application of cybernetics principles.
期刊最新文献
Distributed Predefined-Time Convergent Algorithm for Solving Time-Varying Resource Allocation Problem Over Directed Networks Data-Driven Inverse Reinforcement Learning for Heterogeneous Optimal Robust Formation Control. Predictive Observer-Based Dual-Rate Prescribed Performance Control for Visual Servoing of Robot Manipulators With View Constraints. Observer-Based Control of Networked Periodic Piecewise Systems With Encoding–Decoding Mechanism Granule Margin-Based Feature Selection in Weighted Neighborhood Systems
×
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