Erdogdu, Murat A.
52 publications
NeurIPS
2025
From Information to Generative Exponent: Learning Rate Induces Phase Transitions in SGD
NeurIPS
2025
When Do Transformers Outperform Feedforward and Recurrent Networks? a Statistical Perspective
NeurIPS
2024
A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal Samplers
NeurIPS
2023
Learning in the Presence of Low-Dimensional Structure: A Spiked Random Matrix Perspective
NeurIPS
2023
Optimal Excess Risk Bounds for Empirical Risk Minimization on $p$-Norm Linear Regression
NeurIPS
2022
Generalization Bounds for Stochastic Gradient Descent via Localized $\varepsilon$-Covers
NeurIPS
2022
High-Dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation
COLT
2022
Mirror Descent Strikes Again: Optimal Stochastic Convex Optimization Under Infinite Noise Variance
NeurIPS
2021
An Analysis of Constant Step Size SGD in the Non-Convex Regime: Asymptotic Normality and Bias
COLT
2021
On the Convergence of Langevin Monte Carlo: The Interplay Between Tail Growth and Smoothness
NeurIPS
2020
On the Ergodicity, Bias and Asymptotic Normality of Randomized Midpoint Sampling Method