ProSelfLC: Progressive Self Label Correction for Training Robust Deep Neural Networks

Xinshao Wang, Yang Hua, Elyor Kodirov, David Clifton, Neil Robertson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

To train robust deep neural networks (DNNs), we systematically study several target modification approaches, which include output regularisation, self and non-self label correction (LC). Two key issues are discovered: (1) Self LC is the most appealing as it exploits its own knowledge and requires no extra models. However, how to automatically decide the trust degree of a learner as training goes is not well answered in the literature? (2) Some methods penalise while the others reward low-entropy predictions, prompting us to ask which one is better? To resolve the first issue, taking two well-accepted propositions–deep neural networks learn meaningful patterns before fitting noise [3] and minimum entropy regularisation principle [10]–we propose a novel end-to-end method named ProSelfLC, which is designed according to learning time and entropy. Specifically, given a data point, we progressively increase trust in its predicted label distribution versus its annotated one if a model has been trained for enough time and the prediction is of low entropy (high confidence). For the second issue, according to ProSelfLC, we empirically prove that it is better to redefine a meaningful low-entropy status and optimise the learner toward it. This serves as a defence of entropy minimisation. We demonstrate the effectiveness of ProSelfLC through extensive experiments in both clean and noisy settings.
Original languageEnglish
Title of host publication2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR): Proceedings
Number of pages10
Publication statusAccepted - 01 Mar 2021

Fingerprint Dive into the research topics of 'ProSelfLC: Progressive Self Label Correction for Training Robust Deep Neural Networks'. Together they form a unique fingerprint.

Cite this