6 papers
Certified Robustness from Approximate Gaussian Mixture Structures in Pretrained Latent Spaces
Konstantinos Emmanouilidis, Tianjiao Ding, Nghia Nguyen +2
Deep learning models are vulnerable to adversarial perturbations, raising important concerns for safety-critical deployment. Empirical defenses can achieve strong robustness in pra…
Shuffling the Data, Stretching the Step-size: Sharper Bias in constant step-size SGD
Konstantinos Emmanouilidis, Emmanouil-Vasileios Vlatakis-Gkaragkounis, Rene Vidal
From adversarial robustness to multi-agent learning, many machine learning tasks can be cast as finite-sum min-max optimization or, more generally, as variational inequality proble…
Beyond Test-Time Memory: State-Space Optimal Control for LLM Reasoning
Peihao Wang, Shan Yang, Xijun Wang +8
Associative memory has long underpinned the design of sequential models. Beyond recall, humans reason by projecting future states and selecting goal-directed actions, a capability…
Disentangling Safe and Unsafe Corruptions via Anisotropy and Locality
Ramchandran Muthukumar, Ambar Pal, Jeremias Sulam +1
State-of-the-art machine learning systems are vulnerable to small perturbations to their input, where ``small'' is defined according to a threat model that assigns a positive threa…
Can Implicit Bias Imply Adversarial Robustness?
Hancheng Min, René Vidal
The implicit bias of gradient-based training algorithms has been considered mostly beneficial as it leads to trained networks that often generalize well. However, Frei et al. (2023…
Certified Robustness against Sparse Adversarial Perturbations via Data Localization
Ambar Pal, René Vidal, Jeremias Sulam
Recent work in adversarial robustness suggests that natural data distributions are localized, i.e., they place high probability in small volume regions of the input space, and that…