When are solutions connected in deep networks? QN Nguyen, P Bréchet, M Mondelli Advances in Neural Information Processing Systems 34, 20956-20969, 2021 | 7 | 2021 |
On connectivity of solutions in deep learning: the role of over-parameterization and feature quality Q Nguyen, P Bréchet, M Mondelli Advances in Neural Information Processing Systems 34, 2021 | 5 | 2021 |
Critical points and convergence analysis of generative deep linear networks trained with Bures-Wasserstein loss P Bréchet, K Papagiannouli, J An, G Montúfar International Conference on Machine Learning, 3106-3147, 2023 | 4 | 2023 |
Convergence of Generative Deep Linear Networks Trained with Bures-Wasserstein Loss P Bréchet, K Papagiannouli, J An, G Montufar | | 2022 |
Informative GANs via Structured Regularization of Optimal Transport P Bréchet, T Wu, T Möllenhoff, D Cremers arXiv preprint arXiv:1912.02160, 2019 | | 2019 |