CMW-Net: an adaptive robust algorithm for sample selection and label correction

Jun Shu; Xiang Yuan; Deyu Meng

doi:10.1093/nsr/nwad084

CMW-Net: an adaptive robust algorithm for sample selection and label correction

Natl Sci Rev. 2023 Mar 25;10(6):nwad084. doi: 10.1093/nsr/nwad084. eCollection 2023 Jun.

Authors

Jun Shu^{1

2}, Xiang Yuan^{1

2}, Deyu Meng^{1

2

3

4}

Affiliations

¹ School of Mathematics and Statistics, Xi'an Jiaotong University, China.
² Ministry of Education Key Lab of Intelligent Networks and Network Security, Xi'an Jiaotong University, China.
³ Peng Cheng Laboratory, China.
⁴ Pazhou Laboratory (Huangpu), China.

Abstract

A class-aware sample weighting algorithm is developed for general label noise problems. The algorithm can effectively tackle complicated and diverse noisy label tasks, winning the Championship of the 'Arena Contest' Track 1 of 2022 Greater BayArea (Huangpu) International Algorithm Case Competition.