Research map / Optimization / Nonconvex and smooth optimization
On Faster Convergence of Cyclic Block Coordinate Descent-type Methods for Strongly Convex Minimization
Xingguo Li, Tuo Zhao, Raman Arora, Han Liu, Mingyi Hong
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 |
|---|---|---|---|
| use block coordinate descent instead of gradient descent | elastic-net and logistic regression · Elastic-net and ridge logistic, 4 covariance settings | objective gap vs passes | Improved |
| use cyclic instead of randomized or permuted order, ill-conditioned Hessian | CBCGD · Settings II and III | objective gap vs passes | Hurt |
| use cyclic instead of permuted order, well-conditioned Hessian | CBCGD · Settings I and IV | objective gap vs passes | Mixed |
| use cyclic instead of randomized order | CBCGD · Setting IV | objective gap vs passes | Improved |
| use permuted instead of cyclic order | PBCGD · All four settings | objective gap vs passes | Improved |
Closest papers by meaning
- Cyclic Block Coordinate Descent With Variance Reduction for Composite Nonconvex Optimization (ICML 2023 · similarity 0.89)
- The Sample Complexity of Gradient Descent in Stochastic Convex Optimization (NeurIPS 2024 · similarity 0.88)
- High-probability complexity bounds for stochastic non-convex minimax optimization (NeurIPS 2024 · similarity 0.88)
- Revisiting the Last-Iterate Convergence of Stochastic Gradient Methods (ICLR 2024 · similarity 0.88)
- Block Acceleration Without Momentum: On Optimal Stepsizes of Block Gradient Descent for Least-Squares (ICML 2024 · similarity 0.88)
- Incremental Gradient Descent with Small Epoch Counts is Surprisingly Slow on Ill-Conditioned Problems (ICML 2025 · similarity 0.88)
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.
