Research map / Optimization / Nonconvex and smooth optimization
Smooth Convex Optimization Using Sub-Zeroth-Order Oracles
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 approximate gradient directions from directional-preference queries in an ellipsoid method | Smooth convex optimization with a directional-preference oracle · Smooth Lipschitz convex functions | query complexity: \tilde{\mathcal{O}}(n^{4}) | Helped |
| Infer derivative signs from comparisons and use approximate-gradient ellipsoid cuts | Smooth convex optimization with a comparator oracle · Smooth Lipschitz convex functions | query complexity: \tilde{\mathcal{O}}(n^{4}) | Helped |
| Estimate gradients from value queries along basis directions and use ellipsoid cuts | Smooth convex optimization with a value oracle · Smooth Lipschitz convex functions | query complexity: \tilde{\mathcal{O}}(n^3) | No clear effect |
| Estimate gradient directions from repeated noisy-value samples | Smooth convex optimization with a noisy-value oracle · Smooth Lipschitz convex functions | sample complexity: \tilde{\mathcal{O}}(n^{13}/\varepsilon^{4}) | Hurt |
| Use staged gradient estimation and repeatedly query a point selected for low regret | Regret-NV · Noisy-value oracle, regret over T queries | regret: \tilde{\mathcal{O}}( n^{3.75} \radiusofset{C} \sqrt{\beta \sigma} T^{0.75}) | Mixed |
Closest papers by meaning
- Exploiting Higher Order Smoothness in Derivative-free Optimization and Continuous Bandits (NeurIPS 2020 · similarity 0.91)
- The power of first-order smooth optimization for black-box non-smooth problems (ICML 2022 · similarity 0.90)
- Stochastic Zeroth-Order Optimization under Strongly Convexity and Lipschitz Hessian: Minimax Sample Complexity (NeurIPS 2024 · similarity 0.89)
- Acceleration Exists! Optimization Problems When Oracle Can Only Compare Objective Function Values (NeurIPS 2024 · similarity 0.88)
- A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization (ICML 2025 · similarity 0.88)
- A Universal Transfer Theorem for Convex Optimization Algorithms Using Inexact First-order Oracles (ICML 2024 · 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.