Showing 1–13 of 13 results
/ Date/ Name
May 27, 2024A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal SamplersSep 7, 2023Gradient-Based Feature Learning under Structured DataMar 7, 2026Post-Training with Policy Gradients: Optimality and the Base Model BarrierSep 29, 2022Neural Networks Efficiently Learn Low-Dimensional Representations with SGDAug 14, 2024Learning Multi-Index Models with Neural Networks via Mean-Field Langevin DynamicsJun 5, 2025On Fitting Flow Models with Large Sinkhorn CouplingsJun 24, 2024Mean-Field Langevin Dynamics for Signed Measures via a Bilevel ApproachMar 7, 2023Towards a Complete Analysis of Langevin Monte Carlo: Beyond Poincaré InequalityOct 21, 2024Robust Feature Learning for Multi-Index Models in High DimensionsMar 14, 2025When Do Transformers Outperform Feedforward and Recurrent Networks? A Statistical PerspectiveSep 29, 2025Flow Matching with Semidiscrete CouplingsOct 23, 2025From Information to Generative Exponent: Learning Rate Induces Phase Transitions in SGDApr 21, 2026Super Apriel: One Checkpoint, Many Speeds