Research map / Optimization / Nonconvex and smooth optimization
Adaptive Proximal Average Approximation for Composite Convex Minimization
Li Shen, Wei Liu, Junzhou Huang, Yu-Gang Jiang, Shiqian Ma
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 |
|---|---|---|---|
| replace PA-APG with APA-APG1 | PA-APG · Overlapping group Lasso K=40, error 1e-6 | iterations to reach precision: 18087 -> 653 | Improved |
| replace PA-APG with APA-APG2 | PA-APG · Overlapping group Lasso K=40, error 1e-6 | iterations to reach precision: 18087 -> 1031 | Improved |
| replace PA-APG with APA-APG1 | PA-APG · Overlapping group Lasso K=10, error 1e-4 | iterations to reach precision: 124 -> 25 | Improved |
| replace PA-APG with APA-APG1 | PA-APG · Overlapping group Lasso K=20, error 1e-6 | iterations to reach precision: 5765 -> 76 | Improved |
| increase number of groups K from 10 to 40 at error 1e-6 | PA-APG · Overlapping group Lasso | iterations to reach precision: 1524 -> 18087 | Hurt |
| increase number of groups K from 10 to 40 at error 1e-6 | APA-APG2 · Overlapping group Lasso | iterations to reach precision: 41 -> 1031 | Hurt |
| tighten precision from 1e-4 to 1e-6 | PA-APG · Overlapping group Lasso K=40 | iterations to reach precision: 1531 -> 18087 | Hurt |
| replace PA-APG with APA-APG2 | PA-APG · Graph-guided fused Lasso d=500, error 1e-6 | iterations to reach precision: 14624 -> 465 | Improved |
| replace PA-APG with APA-APG2 | PA-APG · Graph-guided fused Lasso d=2000, error 1e-6 | iterations to reach precision: 50472 -> 4878 | Improved |
| replace PA-APG with APA-APG2 | PA-APG · Graph-guided fused Lasso d=1000, error 1e-6 | iterations to reach precision: 44364 -> 889 | Improved |
| use APA-APG2 instead of APA-APG1 | APA-APG1 · Graph-guided fused Lasso d=500, error 1e-6 | iterations to reach precision: 901 -> 465 | Improved |
| use APA-APG2 instead of APA-APG1 | APA-APG1 · Overlapping group Lasso K=10 and K=20 | iterations to reach precision: 25 -> 25 | No clear effect |
Closest papers by meaning
- On Acceleration for Convex Composite Minimization with Noise-Corrupted Gradients and Approximate Proximal Mapping (JMLR 2022 · similarity 0.89)
- Adaptive Proximal Average Based Variance Reducing Stochastic Methods for Optimization with Composite Regularization (AAAI 2019 · similarity 0.88)
- Extensions to the Proximal Distance Method of Constrained Optimization (JMLR 2022 · similarity 0.88)
- Accelerated Quasi-Newton Proximal Extragradient: Faster Rate for Smooth Convex Optimization (NeurIPS 2023 · similarity 0.88)
- Adaptive Proximal Gradient Method for Convex Optimization (NeurIPS 2024 · similarity 0.87)
- Non-convex Stochastic Composite Optimization with Polyak Momentum (ICML 2024 · 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.
