Picture for Fabian Pedregosa

Fabian Pedregosa

INRIA Saclay - Ile de France

Stepping on the Edge: Curvature Aware Learning Rate Tuners

Add code
Jul 08, 2024
Viaarxiv icon

Are we making progress in unlearning? Findings from the first NeurIPS unlearning competition

Add code
Jun 13, 2024
Viaarxiv icon

Stability-Aware Training of Neural Network Interatomic Potentials with Differentiable Boltzmann Estimators

Add code
Feb 21, 2024
Figure 1 for Stability-Aware Training of Neural Network Interatomic Potentials with Differentiable Boltzmann Estimators
Figure 2 for Stability-Aware Training of Neural Network Interatomic Potentials with Differentiable Boltzmann Estimators
Figure 3 for Stability-Aware Training of Neural Network Interatomic Potentials with Differentiable Boltzmann Estimators
Figure 4 for Stability-Aware Training of Neural Network Interatomic Potentials with Differentiable Boltzmann Estimators
Viaarxiv icon

On the Interplay Between Stepsize Tuning and Progressive Sharpening

Add code
Dec 07, 2023
Viaarxiv icon

On Implicit Bias in Overparameterized Bilevel Optimization

Add code
Dec 28, 2022
Figure 1 for On Implicit Bias in Overparameterized Bilevel Optimization
Figure 2 for On Implicit Bias in Overparameterized Bilevel Optimization
Figure 3 for On Implicit Bias in Overparameterized Bilevel Optimization
Figure 4 for On Implicit Bias in Overparameterized Bilevel Optimization
Viaarxiv icon

A Novel Stochastic Gradient Descent Algorithm for Learning Principal Subspaces

Add code
Dec 08, 2022
Figure 1 for A Novel Stochastic Gradient Descent Algorithm for Learning Principal Subspaces
Figure 2 for A Novel Stochastic Gradient Descent Algorithm for Learning Principal Subspaces
Figure 3 for A Novel Stochastic Gradient Descent Algorithm for Learning Principal Subspaces
Figure 4 for A Novel Stochastic Gradient Descent Algorithm for Learning Principal Subspaces
Viaarxiv icon

Extragradient with Positive Momentum is Optimal for Games with Cross-Shaped Jacobian Spectrum

Add code
Nov 09, 2022
Figure 1 for Extragradient with Positive Momentum is Optimal for Games with Cross-Shaped Jacobian Spectrum
Figure 2 for Extragradient with Positive Momentum is Optimal for Games with Cross-Shaped Jacobian Spectrum
Figure 3 for Extragradient with Positive Momentum is Optimal for Games with Cross-Shaped Jacobian Spectrum
Viaarxiv icon

Second-order regression models exhibit progressive sharpening to the edge of stability

Add code
Oct 10, 2022
Figure 1 for Second-order regression models exhibit progressive sharpening to the edge of stability
Figure 2 for Second-order regression models exhibit progressive sharpening to the edge of stability
Figure 3 for Second-order regression models exhibit progressive sharpening to the edge of stability
Figure 4 for Second-order regression models exhibit progressive sharpening to the edge of stability
Viaarxiv icon

The Curse of Unrolling: Rate of Differentiating Through Optimization

Add code
Sep 27, 2022
Figure 1 for The Curse of Unrolling: Rate of Differentiating Through Optimization
Figure 2 for The Curse of Unrolling: Rate of Differentiating Through Optimization
Figure 3 for The Curse of Unrolling: Rate of Differentiating Through Optimization
Figure 4 for The Curse of Unrolling: Rate of Differentiating Through Optimization
Viaarxiv icon

Only Tails Matter: Average-Case Universality and Robustness in the Convex Regime

Add code
Jun 22, 2022
Figure 1 for Only Tails Matter: Average-Case Universality and Robustness in the Convex Regime
Figure 2 for Only Tails Matter: Average-Case Universality and Robustness in the Convex Regime
Figure 3 for Only Tails Matter: Average-Case Universality and Robustness in the Convex Regime
Figure 4 for Only Tails Matter: Average-Case Universality and Robustness in the Convex Regime
Viaarxiv icon