arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.17628stat.MLcs.AIcs.LGstat.ME
收藏

rSDNet:统一的鲁棒神经学习以对抗标签噪声和对抗攻击

rSDNet: Unified Robust Neural Learning against Label Noise and Adversarial Attacks

Indian Statistical Institute, Kolkata, India(印度统计研究所,加尔各答,印度)

Suryasis Jana, Abhik Ghosh

AI总结:

本文提出rSDNet,通过最小化分歧估计问题,提升神经网络在标签噪声和对抗攻击下的鲁棒性,同时保持清洁数据上的准确率。

摘要:

神经网络是现代人工智能的核心,但其训练对数据污染高度敏感。标准神经分类器通过最小化分类交叉熵损失进行训练,对应于多元模型下的最大似然估计。尽管在理想条件下统计效率高,但易受污染观测的影响,包括标签噪声和对抗扰动。本文提出一种统一且统计上稳健的框架,通过最小化分歧估计问题,引入基于S-分歧类别的rSDNet算法。该方法通过模型概率自动降低异常观测的权重。我们建立了rSDNet在总体层面的性质,包括Fisher一致性、分类校准暗示贝叶斯最优性以及在均匀标签噪声和微小特征污染下的稳健性保证。在三个基准图像分类数据集上的实验表明,rSDNet在提升对抗攻击和标签污染的鲁棒性的同时,保持了清洁数据上的竞争力。

英文摘要:

Neural networks are central to modern artificial intelligence, yet their training remains highly sensitive to data contamination. Standard neural classifiers are trained by minimizing the categorical cross-entropy loss, corresponding to maximum likelihood estimation under a multinomial model. While statistically efficient under ideal conditions, this approach is highly vulnerable to contaminated observations including label noises corrupting supervision in the output space, and adversarial perturbations inducing worst-case deviations in the input space. In this paper, we propose a unified and statistically grounded framework for robust neural classification that addresses both forms of contamination within a single learning objective. We formulate neural network training as a minimum-divergence estimation problem and introduce rSDNet, a robust learning algorithm based on the general class of $S$-divergences. The resulting training objective inherits robustness properties from classical statistical estimation, automatically down-weighting aberrant observations through model probabilities. We establish essential population-level properties of rSDNet, including Fisher consistency, classification calibration implying Bayes optimality, and robustness guarantees under uniform label noise and infinitesimal feature contamination. Experiments on three benchmark image classification datasets show that rSDNet improves robustness to label corruption and adversarial attacks while maintaining competitive accuracy on clean data, Our results highlight minimum-divergence learning as a principled and effective framework for robust neural classification under heterogeneous data contamination.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑