Adversarial robustness of neural networks from the perspective of Lipschitz calculus: A survey
We survey the adversarial robustness of neural networks from the perspective of Lipschitz
calculus in a unifying fashion by expressing models, attacks and safety guarantees, that is, a …
calculus in a unifying fashion by expressing models, attacks and safety guarantees, that is, a …
Prompt certified machine unlearning with randomized gradient smoothing and quantization
The right to be forgotten calls for efficient machine unlearning techniques that make trained
machine learning models forget a cohort of data. The combination of training and unlearning …
machine learning models forget a cohort of data. The combination of training and unlearning …
Expressive 1-lipschitz neural networks for robust multiple graph learning against adversarial attacks
Recent findings have shown multiple graph learning models, such as graph classification
and graph matching, are highly vulnerable to adversarial attacks, ie small input …
and graph matching, are highly vulnerable to adversarial attacks, ie small input …
Robust implicit networks via non-Euclidean contractions
S Jafarpour, A Davydov… - Advances in Neural …, 2021 - proceedings.neurips.cc
Implicit neural networks, aka, deep equilibrium networks, are a class of implicit-depth
learning models where function evaluation is performed by solving a fixed point equation …
learning models where function evaluation is performed by solving a fixed point equation …
CLIP: Cheap Lipschitz training of neural networks
Despite the large success of deep neural networks (DNN) in recent years, most neural
networks still lack mathematical guarantees in terms of stability. For instance, DNNs are …
networks still lack mathematical guarantees in terms of stability. For instance, DNNs are …
Lipschitz regularization for softening material models: the Lip-field approach
N Moës, N Chevaugeon - Comptes …, 2021 - comptes-rendus.academie-sciences …
Softening material models are known to trigger spurious localizations. This may be shown
theoretically by the existence of solutions with zero dissipation when localization occurs and …
theoretically by the existence of solutions with zero dissipation when localization occurs and …
Certified robustness via locally biased randomized smoothing
BG Anderson, S Sojoudi - Learning for Dynamics and …, 2022 - proceedings.mlr.press
The successful incorporation of machine learning models into safety-critical control systems
requires rigorous robustness guarantees. Randomized smoothing remains one of the state …
requires rigorous robustness guarantees. Randomized smoothing remains one of the state …
A quantitative geometric approach to neural-network smoothness
Fast and precise Lipschitz constant estimation of neural networks is an important task for
deep learning. Researchers have recently found an intrinsic trade-off between the accuracy …
deep learning. Researchers have recently found an intrinsic trade-off between the accuracy …
Improving the robustness of transformer-based large language models with dynamic attention
Transformer-based models, such as BERT and GPT, have been widely adopted in natural
language processing (NLP) due to their exceptional performance. However, recent studies …
language processing (NLP) due to their exceptional performance. However, recent studies …
Learning globally smooth functions on manifolds
Smoothness and low dimensional structures play central roles in improving generalization
and stability in learning and statistics. This work combines techniques from semi-infinite …
and stability in learning and statistics. This work combines techniques from semi-infinite …