High-Probability Guarantees for SGD under beta-Heavy-Tailed Gradient Noise

arXiv:2609.32195 · stat.ML, cs.LG · Submitted 2026-09-26 · Read on arXiv

stat.ML, cs.LG

Submitted: 2026-09-26

Updated: 2026-09-26

Terminology

• Generalization — 19× in this paper · explained in 13 episodes
• Homogeneity — 5× in this paper · explained in 2 episodes
• Convergence Guarantees — 4× in this paper · explained in 3 episodes
• Adam — 2× in this paper · explained in 2 episodes
• Loss Function — 2× in this paper · explained in 2 episodes
• Stochastic Gradient Descent (SGD) — 2× in this paper · explained in 5 episodes
• Algorithmic Stability — 1× in this paper · explained in 1 episode
• Generalization Bounds — 1× in this paper · explained in 2 episodes

Sources

• Adam: A Method for Stochastic Optimization
• Revisiting the Last-Iterate Convergence of Stochastic Gradient Methods
• Nonconvex Stochastic Optimization under Heavy-Tailed Noises: Optimal Convergence without Gradient Clipping

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• One Permutation Is All You Need: Fast, Deterministic Feature Importance and Model Stress-Testing
• Online Conformal Prediction for Non-Exchangeable Panel Data
• Deep Time-Series Forecasting in 10 Years: A Survey