让神经网络更抗扰动，通过几何感知的平滑训练提升鲁棒性
TASER: Task-Aware Stein Regularisation for Geometry-Driven Robustness
- 基于斯坦因算子设计新正则化，引导模型与数据分布对齐
- 在CIFAR-10上提升对抗鲁棒性，且不牺牲干净准确率
- 适用于视觉和回归任务，特别适合追求模型稳定性的研究者
现代深度网络在分布偏移和对抗扰动下仍显脆弱，常因输入敏感性过强或结构不当。我们提出TASER（任务感知斯坦因正则化），一种基于朗之万斯坦因算子的训练阶段正则化框架。通过惩罚训练分布下的点态斯坦因残差，TASER促使预测器与数据密度之间实现几何兼容，诱导各向异性、数据感知的平滑性。我们建立了斯坦因正则化与一阶分布偏移敏感性降低之间的理论联系，开发了适配现代架构的可扩展实现变体，并在回归与视觉基准上验证了更强的鲁棒性与稳定性。在CIFAR-10实验中，TASER持续提升主流训练方法的对抗鲁棒性，且未造成统计上显著的干净准确率下降。
原文摘要 · Abstract (English)
Modern deep networks remain fragile under distribution shift and adversarial perturbations, often due to excessive or poorly structured input sensitivity. We introduce TASER (Task-Aware Stein Regularisation), a training-time regularisation framework derived from Langevin Stein operators. By penalising pointwise Stein residuals under the training distribution, TASER encourages geometric compatibility between predictors and data density, inducing anisotropic, data-aware smoothness. We provide theoretical links between Stein regularisation and reduced first-order shift sensitivity, develop scalable implementation variants compatible with modern architectures, and demonstrate improved robustness and stability across regression and vision benchmarks. Across CIFAR-10 experiments, TASER consistently improves the adversarial robustness of established training methods without incurring statistically significant clean-accuracy degradation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。