Alert button
Picture for Peter L. Bartlett

Peter L. Bartlett

Alert button

Large Stepsize Gradient Descent for Logistic Loss: Non-Monotonicity of the Loss Improves Optimization Efficiency

Feb 24, 2024
Jingfeng Wu, Peter L. Bartlett, Matus Telgarsky, Bin Yu

Viaarxiv icon

A Statistical Analysis of Wasserstein Autoencoders for Intrinsically Low-dimensional Data

Feb 24, 2024
Saptarshi Chakraborty, Peter L. Bartlett

Viaarxiv icon

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

Feb 22, 2024
Ruiqi Zhang, Jingfeng Wu, Peter L. Bartlett

Viaarxiv icon

On the Statistical Properties of Generative Adversarial Models for Low Intrinsic Data Dimension

Jan 28, 2024
Saptarshi Chakraborty, Peter L. Bartlett

Viaarxiv icon

How Many Pretraining Tasks Are Needed for In-Context Learning of Linear Regression?

Oct 12, 2023
Jingfeng Wu, Difan Zou, Zixiang Chen, Vladimir Braverman, Quanquan Gu, Peter L. Bartlett

Viaarxiv icon

Sharpness-Aware Minimization and the Edge of Stability

Sep 29, 2023
Philip M. Long, Peter L. Bartlett

Figure 1 for Sharpness-Aware Minimization and the Edge of Stability
Figure 2 for Sharpness-Aware Minimization and the Edge of Stability
Figure 3 for Sharpness-Aware Minimization and the Edge of Stability
Figure 4 for Sharpness-Aware Minimization and the Edge of Stability
Viaarxiv icon

Trained Transformers Learn Linear Models In-Context

Jun 16, 2023
Ruiqi Zhang, Spencer Frei, Peter L. Bartlett

Figure 1 for Trained Transformers Learn Linear Models In-Context
Viaarxiv icon

Prediction, Learning, Uniform Convergence, and Scale-sensitive Dimensions

Apr 24, 2023
Peter L. Bartlett, Philip M. Long

Figure 1 for Prediction, Learning, Uniform Convergence, and Scale-sensitive Dimensions
Viaarxiv icon

Benign Overfitting in Linear Classifiers and Leaky ReLU Networks from KKT Conditions for Margin Maximization

Mar 02, 2023
Spencer Frei, Gal Vardi, Peter L. Bartlett, Nathan Srebro

Viaarxiv icon