arXiv · 2611.14046
Structured Confidence Perturbation via Fisher Geometry for Adversarial Training
Abstract
Adversarial training is often plagued by "pseudo-attacks", perturbations that significantly increase the loss value but fail to truly alter the model's confidence structure. To address this, we propose two complementary approaches to directly attack and defend against confidence structure deformation. First, building upon the traditional loss maximization perspective, we introduce a hybrid objective that integrates cross-entropy and KL divergence, encouraging adversarial examples to simultaneously disrupt prediction accuracy and distribution consistency. Second, we geometrically reconstruct adversarial training by generating perturbations that maximize the distribution distance on the confidence manifold. Theoretical analysis shows that the optimal perturbation direction aligns with the principal eigenvectors of the Fisher information matrix. We also propose an efficient algorithm to avoid high-dimensional matrix computations. Experiments on multiple datasets and model architectures demonstrate that our method outperforms existing methods in adversarial robustness and robust transferability, particularly excelling in structural perturbations within high-confidence regions.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Chenlong Wei, Hanling Wang, Chong Zhang, Qinkai Yu, Fangyu Wu, Xiaobo Jin. 2026-11-10. Structured Confidence Perturbation via Fisher Geometry for Adversarial Training. https://searcharxiv.com/abs/2611.14046
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.