Research map / Optimization / Nonconvex and smooth optimization
An Optimal Structured Zeroth-order Algorithm for Non-smooth Optimization
Marco Rando, Cesare Molinari, Lorenzo Rosasco, Silvia Villa
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 |
|---|---|---|---|
| increase number of orthogonal directions per iteration from 1 to larger values (ablation) | O-ZD (Algorithm 1) · smooth convex synthetic function, d=50, 4000 evaluations | f(x_k) - f(x*) | Improved |
| increase number of orthogonal directions per iteration from 1 to larger values (ablation) | O-ZD (Algorithm 1) · non-smooth convex synthetic function, d=50, 4000 evaluations | f(x_k) - f(x*) | Improved |
| use structured orthogonal directions instead of random Gaussian or spherical directions | finite-difference zeroth-order methods · smooth convex target, d=10, 1000 evaluations | f(x_k) - f(x*) | Improved |
| use structured orthogonal directions instead of random Gaussian or spherical directions | finite-difference zeroth-order methods · non-smooth convex target, d=10, 1000 evaluations | f(x_k) - f(x*) | Improved |
| use larger step-size constant c with Gaussian directions (ablation) | finite-difference with Gaussian directions · smooth convex target | stability of f(x_k) - f(x*) | Hurt |
| use smaller step-size constant c for multiple Gaussian directions (ablation) | multi Gaussian finite difference · non-smooth convex target | f(x_k) - f(x*) | Improved |
Closest papers by meaning
- Decentralized Gradient-Free Methods for Stochastic Non-smooth Non-convex Optimization (AAAI 2024 · similarity 0.88)
- Zeroth-Order Methods for Constrained Nonconvex Nonsmooth Stochastic Optimization (ICML 2024 · similarity 0.88)
- Adaptively Incorporating Directional Hints into Zeroth-Order Optimization (NeurIPS 2026 · similarity 0.88)
- Revisiting Frank-Wolfe for Structured Nonconvex Optimization (NeurIPS 2025 · similarity 0.87)
- Single-Loop Stochastic Algorithms for Difference of Max-Structured Weakly Convex Functions (NeurIPS 2024 · similarity 0.87)
- Efficiently Escaping Saddle Points under Generalized Smoothness via Self-Bounding Regularity (NeurIPS 2025 · similarity 0.87)
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.
