← Search

Mohammed Amine Bennouna

2 accepted papers

2023

Certified Robust Neural Networks: Generalization and Corruption Resistance

ICML 2023poster

Recent work have demonstrated that robustness (to "corruption") can be at odds with generalization. Adversarial training, for instance, aims to reduce the problematic susceptibility of modern neural networks to small data perturbations. Surprisingly, overfitting is a major concern in adversarial tra…

2022

Shallow and Deep Networks are Near-Optimal Approximators of Korobov Functions

ICLR 2022poster

In this paper, we analyze the number of neurons and training parameters that a neural network needs to approximate multivariate functions of bounded second mixed derivatives --- Korobov functions. We prove upper bounds on these quantities for shallow and deep neural networks, drastically lessening t…

Cited by 14SourcePDFScholar