Robust Learning with Adversarial Perturbations and Label Noise
Authors:
Open Access
- 13 December 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Abstract
Despite great success achieved, deep learning methods are vulnerable to noise in the training dataset, including adversarial perturbations and annotation noise. These harmful factors significantly influence the learning process of deep models, leading to less confident models. However, existing methods have not yet studied this practical and challenging issue. In this paper, we propose a novel robust learning method, i.e., Two-Pronged Defense (TPD), which is capable of eliminating negative effects of both data perturbations and label noise during the learning process. On the one hand, to defend against delusive adversarial examples, the proposed method designs an asymmetric adversarial contrastive learning strategy to craft worse-case noisy example for original training data, and train the model to align the semantic between the perturbed data and the original data. In light of this, the TPD would be able to improve the generalization ability of the model on the potential adversarial examples. On the other hand, to combat noisy labels, the TPD applies semi-supervised learning by identifying and discarding noisy labels via a novel designed identification method. Extensive experiments on benchmarks demonstrate the incapability of existing methods and the effectiveness of the proposed method when facing both data and label noise. This work is the very first attempt in learning with data and label noise, and we hope it can pave the way for future studies in related fields.Keywords
Funding Information
- Australian Research Council Discovery Project (DP190102353, CE200100025)
- China Scholarship Council
This publication has 19 references indexed in Scilit:
- Privacy Protection in Deep Multi-modal Retrieval2021
- High-order nonlocal Hashing for unsupervised cross-modal retrievalWorld Wide Web, 2021
- Aggregation-Based Graph Convolutional Hashing for Unsupervised Cross-Modal RetrievalIEEE Transactions on Multimedia, 2021
- Supervised Hierarchical Deep Hashing for Cross-Modal Retrieval2020
- Rethinking Generative Zero-Shot Learning: An Ensemble Learning Perspective for Recognising Visual Patches2020
- CANZSL: Cycle-Consistent Adversarial Networks for Zero-Shot Learning from Natural Language2020
- Symmetric Cross Entropy for Robust Learning With Noisy Labels2019
- Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning2018
- Making Deep Neural Networks Robust to Label Noise: A Loss Correction Approach2017
- Deep Residual Learning for Image Recognition2016