Abstract
The transition matrix, denoting the transition relationship from clean
labels to noisy labels, is essential to build statistically consistent
classifiers in label-noise learning. Existing methods for estimating the
transition matrix rely heavily on estimating the noisy class posterior.
However, the estimation error for noisy class posterior could be large
because of the randomness of label noise. The estimation error would
lead the transition matrix to be poorly estimated. Therefore in this
paper, we aim to solve this problem by exploiting the divide-and-conquer
paradigm. Specifically, we introduce an intermediate class to avoid
directly estimating the noisy class posterior. By this intermediate
class, the original transition matrix can then be factorized into the
product of two easy-to-estimated transition matrices. We term the
proposed method as the dual T-estimator. Both theoretical analyses and empirical results illustrate the effectiveness of the dual T-estimator for estimating transition matrices, leading to better classification performances.
Original language | English |
---|---|
Title of host publication | 34th Conference on Neural Information Processing Systems (NeurIPS 2020) |
Editors | H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, H. Lin |
Publisher | Neural Information Processing Systems Foundation |
Pages | 7260-7271 |
Number of pages | 12 |
Volume | 9 |
ISBN (Print) | 9781713829546 |
DOIs | |
Publication status | Published - 6 Dec 2020 |
Event | 34th Conference on Neural Information Processing Systems, NeurIPS 2020 - Virtual, Online Duration: 6 Dec 2020 → 12 Dec 2020 https://neurips.cc/Conferences/2020 https://proceedings.neurips.cc/paper/2020 |
Publication series
Name | Advances in Neural Information Processing Systems |
---|---|
Volume | 33 |
ISSN (Print) | 1049-5258 |
Name | NeurIPS Proceedings |
---|
Conference
Conference | 34th Conference on Neural Information Processing Systems, NeurIPS 2020 |
---|---|
Period | 6/12/20 → 12/12/20 |
Internet address |
Scopus Subject Areas
- Computer Networks and Communications
- Information Systems
- Signal Processing