Nadav Cohen - Do Neural Nets Need Gradient Descent to Generalize?

RoMaDS: A day on Random Matrix Theory and Deep Learning, September 25, 2025

Nadav Cohen (Tel Aviv University)
Do Neural Nets Need Gradient Descent to Generalize? Matrix Factorization as a Theoretical Testbed

Learn more about RoMaDS at https://www.mat.uniroma2.it/~rds/abou...