I am a postdoc fellow at the Simons Institute at UC Berkeley hosted by Peter Bartlett and Bin Yu. I am a part of the NSF/Simons Collaboration on the Theoretical Foundations of Deep Learning. I received my Ph.D. in Computer Science at Johns Hopkins University, advised by Vladimir Braverman.
I work on the theory and algorithms for machine learning. I am interested in deep learning theory, optimization, and statistical learning.
Jingfeng Wu is a postdoctoral fellow at Simons Institute at UC Berkeley. His research focuses on deep learning theory, optimization, and statistical learning. Prior to that, he earned his PhD from Johns Hopkins University, and MS and BS from Peking University.
Large Stepsizes Accelerate Gradient
Descent for Regularized Logistic Regression
NeurIPS 2025
Minimax Optimal Convergence of Gradient
Descent in Logistic Regression via Large and Adaptive Stepsizes
ICML 2025 (journal version in JMLR 2026)
Implicit Bias of Gradient Descent for
Logistic Regression at the Edge of Stability
NeurIPS 2023 (spotlight)
Risk Comparisons in Linear Regression:
Implicit Regularization Dominates Explicit Regularization
COLT 2026
Benefits of Early Stopping in Gradient
Descent for Overparameterized Logistic Regression
ICML 2025
Last Iterate Risk Bounds of SGD with
Decaying Stepsize for Overparameterized Linear Regression
ICML 2022 (long presentation)
Benign Overfitting of Constant-Stepsize SGD for Linear Regression
COLT 2021 (journal version in JMLR 2023)
