preprint

On the Hardness of Robustness Transfer: A Perspective from Rademacher Complexity over Symmetric Difference Hypothesis Space (preprint)
Recent studies demonstrated that the adversarially robust learning under {{</* math />}}$\ell_\infty${{</ /math />}} attack …
Yuyang Deng, Nidham Gazagnadou, Junyuan Hong, Mehrdad Mahdavi, Lingjuan Lyu
PDF Cite
Cutting some slack for SGD with adaptive Polyak stepsizes (preprint)
Tuning the step size of stochastic gradient descent is tedious and error prone. This has motivated the development of methods that …
Robert M Gower, Mathieu Blondel, Nidham Gazagnadou, Fabian Pedregosa
PDF Cite
Expected smoothness for stochastic variance-reduced methods and sketch-and-project methods for structured linear systems (PhD thesis)
The considerable increase in the number of data and features complicates the learning phase requiring the minimization of a loss …
Nidham Gazagnadou
PDF Cite
Expected smoothness for stochastic variance-reduced methods and sketch-and-project methods for structured linear systems (PhD thesis)