English(EN)Over-parameterization and Adversarial Robustness in Neural Networks: An Overview and Empirical Analysis
新研究通过新颖的蒸馏和测试方法来提升AI鲁棒性 · 跟踪8个来源
作者PulseAugur 编辑部·[9 个来源]·
研究人员正在探索增强神经网络对抗鲁棒性的新方法。一种名为AD-CERT的方法结合了对抗蒸馏和区间界传播，在鲁棒性基准测试上取得了最先进的认证性能。另一种技术，激活放大与衰减（A3），使用轻量级插件模块动态重缩放激活，以最小的开销提高鲁棒性。此外，研究还调查了输入维度的作用，发现更高维度通常更容易构建和控制对抗样本。来自Resemble AI的Proteus框架通过系统地应用音频转换来自动化音频深度伪造检测器的对抗鲁棒性测试。
AI
arXiv:2606.31653v1 Announce Type: cross Abstract: Certified training aims to produce models whose predictions can be formally verified against adversarial perturbations, typically by optimising upper bounds on the worst-case loss over an allowed perturbation set. For neural netwo…
Certified training aims to produce models whose predictions can be formally verified against adversarial perturbations, typically by optimising upper bounds on the worst-case loss over an allowed perturbation set. For neural networks, certified training methods based purely on ti…
arXiv cs.AI
TIER_1English(EN)·Nicolas M. M\"uller, Aditya Tirumala Bukkapatnam, Zohaib Ahmed·
arXiv:2606.29544v1 Announce Type: cross Abstract: We present Proteus, a framework developed at Resemble AI for automated robustness testing of our audio deepfake detection system. Given a detector, Proteus systematically searches over sequences of everyday audio transformations (…
arXiv:2606.27784v1 Announce Type: cross Abstract: The existence of adversarial attacks is often attributed to the presence of non-robust features in neural networks. While prior defenses reduce their impact via pruning, masking, or feature recalibration, we instead propose to joi…
The existence of adversarial attacks is often attributed to the presence of non-robust features in neural networks. While prior defenses reduce their impact via pruning, masking, or feature recalibration, we instead propose to jointly learn to amplify and attenuate these signals …
arXiv:2406.10090v3 Announce Type: replace Abstract: Thanks to their extensive capacity, over-parameterized neural networks exhibit superior predictive capabilities and generalization. However, having a large parameter space is considered one of the main suspects of the neural net…
arXiv cs.LG
TIER_1English(EN)·Matthew J. Colbrook, Igor Mezi\'c, Alexei Stepanenko·
arXiv:2407.06312v2 Announce Type: replace-cross Abstract: Many systems resist analytical modeling, making data-driven inference of dynamics important. Yet data-driven methods can fail to converge or generalize, leaving open a central question: When can system behavior be learned …
arXiv:2606.26207v1 Announce Type: new Abstract: Several theoretical works have tried to explain the adversarial vulnerability of deep neural networks through properties of high-dimensional geometry. However, the assumptions underlying these works are rarely examined empirically, …
Several theoretical works have tried to explain the adversarial vulnerability of deep neural networks through properties of high-dimensional geometry. However, the assumptions underlying these works are rarely examined empirically, and systematic evidence remains limited. In this…
