Research map / Optimization / Nonconvex and smooth optimization
High-Probability Bounds for the Last Iterate of Clipped SGD
Savelii Chezhegov, Daniela Angela Parletta, Andrea Paudice, Eduard Gorbunov
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 |
|---|---|---|---|
| evaluate last iterate instead of average iterate of Clipped-SGD | Clipped-SGD with theory schedules · Pareto-noise logistic objective (d=100) | function value (0.95-percentile and mean) | Improved |
| evaluate last iterate instead of average iterate of Clipped-SGD | Clipped-SGD with theory schedules · Pareto-noise quadratic 0.5||x||^2 (d=100) | function value | Improved |
| evaluate last iterate instead of average iterate of Clipped-SGD | Clipped-SGD with theory schedules · Heavy-tailed logistic regression (d=10, Student-t 2.001 dof) | function value | Improved |
Closest papers by meaning
- Convergence of Clipped SGD on Convex $(L_0,L_1)$-Smooth Functions (NeurIPS 2025 · similarity 0.92)
- Revisiting the Last-Iterate Convergence of Stochastic Gradient Methods (ICLR 2024 · similarity 0.92)
- Clipped SGD Algorithms for Performative Prediction: Tight Bounds for Stochastic Bias and Remedies (ICML 2025 · similarity 0.90)
- Fast Last-Iterate Convergence of SGD in the Smooth Interpolation Regime (NeurIPS 2025 · similarity 0.90)
- The Sample Complexity of Gradient Descent in Stochastic Convex Optimization (NeurIPS 2024 · similarity 0.90)
- Near-Optimal Streaming Heavy-Tailed Statistical Estimation with Clipped SGD (NeurIPS 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.
