Adversarial Concurrent Training: Optimizing Robustness and Accuracy Trade-off of Deep Neural Networks
Preprint
Abstract
Adversarial training has been proven to be an effective technique for improving the adversarial robustness of models. However, there seems to be an inherent trade-off between optimizing the model for accuracy and robustness. To this end, we propose Adversarial Concurrent Training (ACT), which employs adversarial training in a collaborative learning framework whereby we train a robust model in conjunction with a natural model in a minimax game. ACT encourages the two models to align their feature space by using the task-specific decision boundaries and explore the input space more broadly. Furthermore, the natural model acts as a regularizer, enforcing priors on features that the robust model should learn. Our analyses on the behavior of the models show that ACT leads to a robust model with lower model complexity, higher information compression in the learned representations, and high posterior entropy solutions indicative of convergence to a flatter minima. We demonstrate the effectiveness of the proposed approach across different datasets and network architectures. On ImageNet, ACT achieves 68.20% standard accuracy and 44.29% robustness accuracy under a 100-iteration untargeted attack, improving upon the standard adversarial training method's 65.70% standard accuracy and 42.36% robustness.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Knowledge Distillation Beyond Model Compression2021
- Entropy-SGD: biasing gradient descent into wide valleys*Journal of Statistical Mechanics: Theory and Experiment, 2019
- A Discussion of 'Adversarial Examples Are Not Bugs, They Are Features'Distill, 2019
- Adversarial Examples in the Physical World2018
- Recent Trends in Deep Learning Based Natural Language Processing [Review Article]IEEE Computational Intelligence Magazine, 2018
- DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks2016
- Wide Residual Networks2016