Follow
Michael E. Sander
Michael E. Sander
Other namesMichael Eli Sander
Google DeepMind
Verified email at google.com - Homepage
Title
Cited by
Cited by
Year
Vision Transformers provably learn spatial structure
S Jelassi, ME Sander, Y Li
Advances in Neural Information Processing Systems 35, 2022
1042022
Sinkformers: Transformers with doubly stochastic attention
ME Sander, P Ablin, M Blondel, G Peyré
International Conference on Artificial Intelligence and Statistics, 2022
872022
Momentum Residual Neural Networks
ME Sander, P Ablin, M Blondel, G Peyré
International Conference on Machine Learning 139, 9276-9287, 2021
742021
Do Residual Neural Networks discretize Neural Ordinary Differential Equations?
ME Sander, P Ablin, G Peyré
Advances in Neural Information Processing Systems 35, 2022
362022
Fast, Differentiable and Sparse Top-k: a Convex Analysis Perspective
ME Sander, J Puigcerver, J Djolonga, G Peyré, M Blondel
International Conference on Machine Learning, 2023
252023
Implicit regularization of deep residual networks towards neural ODEs
P Marion, YH Wu, ME Sander, G Biau
International Conference on Learning Representations, 2024
182024
How do Transformers perform In-Context Autoregressive Learning?
ME Sander, R Giryes, T Suzuki, M Blondel, G Peyré
International Conference on Machine Learning, 2024
152024
Towards understanding the universality of Transformers for next-token prediction
ME Sander, G Peyré
International Conference on Learning Representations, 2025
32025
Joint learning of energy-based models and their partition function
ME Sander, V Roulet, T Liu, M Blondel
arXiv preprint arXiv:2501.18528, 2025
22025
Unveiling the secrets of paintings: deep neural networks trained on high-resolution multispectral images for accurate attribution and authentication
ME Sander, T Sander, M Sylvestre
Sixteenth International Conference on Quality Control by Artificial Vision …, 2023
22023
Loss Functions and Operators Generated by f-Divergences
V Roulet, T Liu, N Vieillard, ME Sander, M Blondel
arXiv preprint arXiv:2501.18537, 2025
2025
Deeper Learning: Residual Networks, Neural Differential Equations and Transformers, in Theory and Action
ME Sander
Ecole Normale Supérieure de Paris-ENS Paris, 2024
2024
The system can't perform the operation now. Try again later.
Articles 1–12