A Topological Filter for Learning with Label Noise
Pengxiang Wu, Songzhu Zheng, Mayank Goswami, Dimitris Metaxas, Chao Chen
Abstract
Noisy labels can impair the performance of deep neural networks. To tackle this problem, in this paper, we propose a new method for filtering label noise. Unlike most existing methods relying on the posterior probability of a noisy classifier, we focus on the much richer spatial behavior of data in the latent representational space. By leveraging the high-order topological information of data, we are able to collect most of the clean data and train a high-quality model. Theoretically we prove that this topological approach is guaranteed to collect the clean data with high probability. Empirical results show that our method outperforms the state-of-the-arts and is robust to a broad spectrum of noise types and levels.
BibTeX
@inproceedings{NEURIPS2020_f4e3ce3e,
author = {Wu, Pengxiang and Zheng, Songzhu and Goswami, Mayank and Metaxas, Dimitris and Chen, Chao},
booktitle = {Advances in Neural Information Processing Systems},
editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
pages = {21382--21393},
publisher = {Curran Associates, Inc.},
title = {A Topological Filter for Learning with Label Noise},
url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/f4e3ce3e7b581ff32e40968298ba013d-Paper.pdf},
volume = {33},
year = {2020}
}