Research map / Optimization / Nonconvex and smooth optimization
On Graduated Optimization for Stochastic Non-Convex Problems
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| graduated optimization (GradOpt_G) with decaying smoothing starting at delta=7 instead of plain MSGD | training a 1-hidden-layer NN on down-scaled MNIST · MNIST 10x10 | training objective / final error | Helped |
| running optimization on fixed smoothed objective (delta in {1,3,5,7}) from the MSGD stall point instead of unsmoothed (ablation) | NN stalled at MSGD point after 5e4 iterations · MNIST 10x10 | training error convergence | Mixed |
| gradient-oracle SGD with smoothed gradients (SGO_G) on sigma-nice objectives, O(1/sigma^2 eps^2) bound | stochastic non-convex optimization with noisy gradient feedback · theoretical (sigma-nice functions) | iterations to eps-optimal | No clear effect |
| zero-order variant GradOpt_V using value-only oracle with sphere-sampling smoothed gradients | bandit/noisy value feedback on sigma-nice objectives · theoretical | rounds to eps-optimal | Mixed |
Closest papers by meaning
- Improved Last-Iterate Convergence of Shuffling Gradient Methods for Nonsmooth Convex Optimization (ICML 2025 · similarity 0.86)
- Zeroth-Order Methods for Constrained Nonconvex Nonsmooth Stochastic Optimization (ICML 2024 · similarity 0.86)
- Stochastic Zeroth-Order Optimization under Strongly Convexity and Lipschitz Hessian: Minimax Sample Complexity (NeurIPS 2024 · similarity 0.86)
- Universal Gradient Methods for Stochastic Convex Optimization (ICML 2024 · similarity 0.86)
- Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization (NeurIPS 2024 · similarity 0.86)
- Last-Iterate Convergence of Single-Loop Stochastic Methods for Constrained Convex-Concave Minimax Problems (NeurIPS 2026 · similarity 0.86)
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.