no code implementations • 1 Nov 2021 • Reda Chhaibi, Tariq Daouda, Ezechiel Kahn
We present a reliable and very fast method for computing the associated spectral densities, for given architecture and initialization.
1 code implementation • 16 May 2020 • Tariq Daouda, Reda Chhaibi, Prudencio Tossou, Alexandra-Chloé Villani
This work introduces a geometric framework and a novel network architecture for creating correspondences between samples of different conditions.
no code implementations • 4 Jun 2018 • Tariq Daouda, Jeremie Zumer, Claude Perreault, Sébastien Lemieux
Representation learning is at the heart of what makes deep learning effective.