Pith. sign in

Improved Adversarial Training via Learned Optimizer

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Adversarial attack has recently become a tremendous threat to deep learning models. To improve the robustness of machine learning models, adversarial training, formulated as a minimax optimization problem, has been recognized as one of the most effective defense mechanisms. However, the non-convex and non-concave property poses a great challenge to the minimax training. In this paper, we empirically demonstrate that the commonly used PGD attack may not be optimal for inner maximization, and improved inner optimizer can lead to a more robust model. Then we leverage a learning-to-learn (L2L) framework to train an optimizer with recurrent neural networks, providing update directions and steps adaptively for the inner problem. By co-training optimizer's parameters and model's weights, the proposed framework consistently improves the model robustness over PGD-based adversarial training and TRADES.

citation-role summary

background 1

citation-polarity summary

fields

quant-ph 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

support 1

representative citing papers

citing papers explorer

Showing 1 of 1 citing paper.