| Implicit bias of sgd for diagonal linear networks: a provable benefit of stochasticity S Pesme, L Pillaud-Vivien, N Flammarion Neurips 2021, 2021 | 155 | 2021 |
| (S) GD over Diagonal Linear Networks: Implicit Regularisation, Large Stepsizes and Edge of Stability M Even, S Pesme, S Gunasekar, N Flammarion Neurips 2023, 2023 | 65* | 2023 |
| Saddle-to-Saddle Dynamics in Diagonal Linear Networks S Pesme, N Flammarion Neurips 2023, 2023 | 59 | 2023 |
| Online robust regression via sgd on the l1 loss S Pesme, N Flammarion Neurips 2020, 2020 | 47 | 2020 |
| On convergence-diagnostic based step sizes for stochastic gradient descent S Pesme, A Dieuleveut, N Flammarion ICML 2020, 2020 | 28 | 2020 |
| Leveraging Continuous Time to Understand Momentum When Training Diagonal Linear Networks HG Papazov, S Pesme, N Flammarion AISTATS 2024, 2024 | 13 | 2024 |
| Implicit Bias of Mirror Flow on Separable Data S Pesme, RA Dragomir, N Flammarion Neurips 2024, 2024 | 5 | 2024 |
| Deep learning theory through the lens of diagonal linear networks SW Pesme PhD Thesis, EPFL, 2024 | 4 | 2024 |
| MAP Estimation with Denoisers: Convergence Rates and Guarantees S Pesme, G Meanti, M Arbel, J Mairal Neurips 2025, 2025 | 2 | 2025 |
| A Theoretical Framework for Grokking: Interpolation followed by Riemannian Norm Minimisation E Boursier, S Pesme, RA Dragomir Neurips 2025, 2025 | | 2025 |
Nicolas FlammarionEPFLVerified email at epfl.ch
