Peter L. Bartlett

activity

20232026

most citedIn-Context Learning of a Linear Transformer Block: Benefits of the MLP Component and One-Step GD Initialization

1 citations · 5 across the 15 of their papers we have counts for

collaborators

Showing 2025 · stat.MLShow all

3 papers · 2 filters

stat.ML2025★ 1 cited

Risk Comparisons in Linear Regression: Implicit Regularization Dominates Explicit Regularization

Jingfeng Wu, Peter L. Bartlett, Sham M. Kakade +2

Existing theory suggests that for linear regression problems categorized by capacity and source conditions, gradient descent (GD) is always minimax optimal, while both ridge regres…

stat.ML2025

Large Stepsizes Accelerate Gradient Descent for Regularized Logistic Regression

Jingfeng Wu, Pierre Marion, Peter Bartlett

We study gradient descent (GD) with a constant stepsize for $ℓ_{2}$ -regularized logistic regression with linearly separable data. Classical theory suggests small stepsizes to ensu…

stat.ML2025

Minimax Optimal Convergence of Gradient Descent in Logistic Regression via Large and Adaptive Stepsizes

Ruiqi Zhang, Jingfeng Wu, Licong Lin +1

We study $gradient descent$ (GD) for logistic regression on linearly separable data with stepsizes that adapt to the current risk, scaled by a constant hyperparameter $η$ .…