Research map / Optimization / Nonconvex and smooth optimization
Convergence Rates of Non-Convex Stochastic Gradient Descent Under a Generic Lojasiewicz Condition and Local Smoothness
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 |
|---|---|---|---|
| use a loss whose KL* function is sqrt-like (MSE) instead of 1-e^{-x} type (CE) | SGD training of over-parameterized ResNet-18 · Theory, Table 1 and Section 5.3 | SGD time to precision eps (order) | Helped |
| empirical check: gradient-norm vs loss slope matches predicted KL* lower bound | ResNet-18 on MNIST with MSE, CE, CE2 losses · MNIST experiments (Figure 1) | slope of gradient norm vs loss | No clear effect |
Closest papers by meaning
- Fast Convergence of Natural Gradient Descent for Over-Parameterized Neural Networks (NeurIPS 2019 · similarity 0.91)
- The Global Convergence Time of Stochastic Gradient Descent in Non-Convex Landscapes: Sharp Estimates via Large Deviations (ICML 2025 · similarity 0.91)
- Loss Landscape Characterization of Neural Networks without Over-Parametrization (NeurIPS 2024 · similarity 0.90)
- An Improved Analysis of Training Over-parameterized Deep Neural Networks (NeurIPS 2019 · similarity 0.90)
- Overparameterized Nonlinear Learning: Gradient Descent Takes the Shortest Path? (ICML 2019 · similarity 0.90)
- Global Convergence and Stability of Stochastic Gradient Descent (NeurIPS 2022 · similarity 0.90)
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.