Research map / Optimization / Nonconvex and smooth optimization
A Parameter-Free Conditional Gradient Method for Composite Minimization under Hölder Condition
Masaru Ito, Zhaosong Lu, Chuan He
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 parameter-free adaptive line-search step (Algorithm 2) instead of 2/(t+2) step | conditional gradient method · lp-norm over lq ball (n=5000, q=3, p=1.3) | avg CPU time (s): 37.3 -> 28.4 | Improved |
| use parameter-free step (Algorithm 2) instead of parameter-dependent Algorithm 1 | conditional gradient method · lp-norm over lq ball (n=5000, q=3, p=1.3) | avg CPU time (s): 161.2 -> 28.4 | Improved |
| use parameter-free step (Algorithm 2) instead of Algorithm 1 at p=2 | conditional gradient method · lp-norm over lq ball (n=5000, q=3, p=2) | avg CPU time (s): 0.084 -> 0.083 | No clear effect |
| use Algorithm 2 instead of Nesterov step rule (25) | conditional gradient method · Entropy-regularized lp-norm (m=5000, lambda=10, p=1.5) | avg CPU time (s): 1.77 -> 0.33 | Improved |
| use Algorithm 2 instead of Algorithm 1 | conditional gradient method · Entropy-regularized lp-norm (m=5000, lambda=10, p=1.5) | avg iterations: 107117.5 -> 16.3 | Improved |
| use Algorithm 2 instead of Algorithm 1 at p=2, lambda=10 | conditional gradient method · Entropy-regularized lp-norm (m=5000, lambda=10, p=2) | avg CPU time (s): 0.78 -> 0.33 | No clear effect |
| use Algorithm 2 instead of CGM-LS | conditional gradient method · Simplex-constrained NMF (m=n=500, k=5) | avg CPU time (s): 179.24 -> 14.74 | Improved |
| use Algorithm 2 instead of CGM-LS | conditional gradient method · Simplex-constrained NMF (m=n=500, k=5) | avg iterations: 7131.6 -> 904.3 | Improved |
| increase k from 5 to 10 | Algorithm 2 · Simplex-constrained NMF (m=n=500) | avg CPU time (s): 14.74 -> 6.35 | Improved |
Closest papers by meaning
- Nonlinearly Preconditioned Gradient Methods under Generalized Smoothness (ICML 2025 · similarity 0.88)
- Accelerated Distance-adaptive Methods for Hölder Smooth and Convex Optimization (NeurIPS 2025 · similarity 0.87)
- Decentralized Gradient-Free Methods for Stochastic Non-smooth Non-convex Optimization (AAAI 2024 · similarity 0.87)
- Optimizing $(L_0, L_1)$-Smooth Functions by Gradient Methods (ICLR 2025 · similarity 0.86)
- A Study of First-Order Methods with a Deterministic Relative-Error Gradient Oracle (ICML 2024 · similarity 0.86)
- Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization (NeurIPS 2024 · similarity 0.85)
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.
