科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ IEEE transactions on neural networks and learning systems2026-09-02

Toward Improving Stochastic Neural Network Robustness via Arbitrary Distribution Injection.

Rui Zhou, Hao Yang, Wenxu Wang, Dingjiang Wei, Peng Lin, Cheng Zhu, Yun Zhou

原始摘要(英文原文)· Original abstract
Adversarial attacks pose significant challenges to the security and robustness of deep-learning models. Stochastic neural networks (SNNs) have shown promising effectiveness in improving robustness by injecting stochastic noise into model activations, features, or weights. However, most existing SNN-based defenses rely on predefined distributional forms, such as Gaussian or Uniform. In real-world scenarios, data distributions are often non-Gaussian, skewed, or multimodal, which cannot be adequately captured by such fixed assumptions, thereby limiting the robustness of existing methods. To address this limitation, we propose a novel SNN named arbitrary distribution injection (ADI), which enables distribution modeling from nonpredefined, data-dependent distributions. In particular, we introduce a conditional stochastic feature mapping mechanism to model feature distributions, together with a theoretically grounded variance-regularization loss. Extensive experiments across diverse attack methods, datasets, modalities, and network architectures show that ADI achieves robustness improvements and promising generalization across the evaluated settings. Furthermore, detailed parameter analyses and feature-distribution visualizations provide deeper insights into the underlying mechanisms of ADI.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Toward Improving Stochastic Neural Network Robustness via Arbitrary Distribution Injection. — 科研速览 Science Skim