cs.LGNov 5, 2024

Survival of the Fittest Perturbations: Uncertainty-Aware Evolution of Adversarial Distributions for Robust Generalization

Authors: Junhao Dong, Zicheng Yu, Xinghua Qu, Z. Jane Wang, Yew-Soon Ong

Organizations: College of Computing and Data Science, Nanyang Technological University, Singapore · Center for Frontier AI Research, Agency for Science, Technology and Research (A*STAR), Singapore · Bytedance, Singapore · Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC V6T 1Z4, Canada

Abstract

Despite deep learning's broad successes across diverse scenarios, its vulnerability to adversarial examples impedes deployment, making adversarial training a leading approach to improve robustness. However, pointwise adversarial training maps each clean sample to a single adversary, overlooking surrounding adversarial variability and the associated shifts in feature statistics. To address these issues, we propose Evolutionary Uncertainty-Aware Distributional Adversarial Training (EvoUAD-AT). Specifically, we design an evolutionary adversarial uncertainty modeling module that maintains an instance-wise search distribution over adversarial feature statistics around an adversarially generated anchor. Candidates are evaluated under the current model, selected by their threat to a benign reference, and used to adapt the distribution; strategy parameters are retained across epochs. To avoid anchoring on misclassified clean samples, we refine the alignment reference and match benign and adversarial distributions. Introspective gradient matching further regularizes input gradients without external supervision. Our analysis characterizes the elite fitness surrogate and gives a conditional certificate for latent adversarial features. Extensive evaluations across multiple benchmarks demonstrate that our approach consistently outperforms existing state-of-the-art methods in both clean accuracy and robustness, with thorough analyses confirming the effectiveness of each proposed component.

Figures & tables

Explore similar work

CardsList
  1. Robustness Meets Uncertainty: Evidential Adversarial Training for Robust Selective Classification

    Jul 3, 2026Nicolas Sournac, Ahmed Baha Ben Jmaa, Bertrand BraeckeveldtAdversarial TrainingEvidential Deep Learning

  2. Probabilistic Adversarial Training

    Sep 30, 2026Andi Zhang, Xingyu Zhao, Siddartha KhastgirAdversarial TrainingAdversarial Robustness

  3. Learning Robustness at Test-Time from a Non-Robust Teacher

    Apr 13, 2026Stefano Bianchettin, Giulio Rossolini, Giorgio ButtazzoAdversarial RobustnessTest-Time Adaptation