How to Prevent the Continuous Damage of Noises to Model Training?

Author:

Yu Xiaotian1,Jiang Yang2,Shi Tianqi2,Feng Zunlei1,Wang Yuexuan3,Song Mingli1,Sun Li4

Affiliation:

1. Zhejiang University

2. Alibaba Group

3. University of Hong Kong

4. Ningbo Innovation Center Zhejiang University

Funder

National Key Research and Development Project

Ningbo Natural Science Foundation

Fundamental Research Funds for the Central Universities

Publisher

IEEE

Reference56 articles.

1. Mentornet: Learning data-driven curriculum for very deep neural networks on corrupted labels;jiang;International Conference on Machine Learning,0

2. Imagenet: A large-scale hierarchical image database;jia;Proceedings of the IEEE/CVF International Conference on Computer Vision,0

3. Robust curriculum learning: from clean label detection to noisy label self-correction;zhou;International Conference on Learning Representations,0

4. Amalgamating Knowledge from Heterogeneous Graph Neural Networks

5. Learning graph neural networks for image style transfer;jing;ECCV,2022

Cited by 3 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Variational Rectification Inference for Learning with Noisy Labels;International Journal of Computer Vision;2024-08-12

2. Decoding class dynamics in learning with noisy labels;Pattern Recognition Letters;2024-08

3. Bayesian DivideMix++ for Enhanced Learning with Noisy Labels;Neural Networks;2024-04

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3