对抗训练神经网络显著图稀疏性获理论解释

1. 12:00
   arXiv · Machine Learning Theory

   对抗训练神经网络显著性图稀疏性的理论解释

   论文针对对抗训练神经网络梯度显著性图呈现稀疏性的经验观察，为两层 ReLU 网络给出理论解释。研究基于对抗训练等价于带权重衰减惩罚与对抗全变差项的经验风险最小化，证明在数据点与神经元数量增长、正则化参数按适当速率趋零时，极小化解收敛到具有最小梯度范数与 Barron 范数的 Bayes 分类器。

Current heat 9·Comparable peak 10（Oct 9）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.