Research map / Optimization / Nonconvex and smooth optimization
High Probability Convergence Bounds for Non-convex Stochastic Gradient Descent with Sub-Weibull Noise
Liam Madden, Emiliano Dall'Anese, Stephen Becker
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase noise tail heaviness theta from 2 to 4 | SGD convergence error · Synthetic two-layer GELU net, delta in [0.001,0.01] | fitted exponent of log(1/delta): 1.40 -> 4.08 | Hurt |
| increase noise tail heaviness theta from 2 to 4 | SGD convergence error · Synthetic two-layer GELU net, delta in [0.1,0.2] | fitted exponent of log(1/delta): 0.64 -> 1.51 | Hurt |
| consider smaller failure probability delta ranges | SGD convergence error exponent · Synthetic two-layer GELU net, theta=3 | fitted exponent of log(1/delta): 0.93 -> 1.56 -> 2.33 | Hurt |
Closest papers by meaning
- High Probability Convergence of Stochastic Gradient Methods (ICML 2023 · similarity 0.92)
- High-probability complexity bounds for stochastic non-convex minimax optimization (NeurIPS 2024 · similarity 0.91)
- Revisiting the Last-Iterate Convergence of Stochastic Gradient Methods (ICLR 2024 · similarity 0.90)
- Error Analysis Affected by Heavy-Tailed Gradients for Non-Convex Pairwise Stochastic Gradient Descent (AAAI 2025 · similarity 0.89)
- Statistical Guarantees for High-Dimensional Stochastic Gradient Descent (NeurIPS 2025 · similarity 0.89)
- High-Probability Convergence for Composite and Distributed Stochastic Minimization and Variational Inequalities with Heavy-Tailed Noise (ICML 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
