Research map / Optimization / Nonconvex and smooth optimization
UnderGrad: A Universal Black-Box Optimization Method with Almost Dimension-Free Convergence Rate Guarantees
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 |
|---|---|---|---|
| Replace UNIXGRAD's Bregman-diameter-based step size with UNDERGRAD's adaptive dual-extrapolation step size (reweighted gradients, mirror map) using entropic regularizer on simplex | UNDERGRAD · Simplex (d=100), linear losses, perfect oracle | suboptimality gap f(x)-min f convergence rate | Helped |
| UNIXGRAD with small initial step size (10E-3 or smaller) versus UNDERGRAD step size (ablation) | UNIXGRAD · Simplex (d=100), perfect oracle | convergence rate | Mixed |
| UNIXGRAD with initial step size equal to or larger than UNDERGRAD's learning rate (ablation) | UNIXGRAD · Simplex (d=100), perfect oracle | convergence rate | Hurt |
| Gradient reweighting with UNDERGRAD step size on noisy oracle | UNDERGRAD · Simplex (d=100), noisy SFO | convergence rate | Helped |
| UNIXGRAD with small initial step size (e.g., 10^-3 times eta_1) in noisy setting (ablation) | UNIXGRAD · Simplex (d=100), noisy SFO | convergence rate | Hurt |
Closest papers by meaning
- Universality of AdaGrad Stepsizes for Stochastic Optimization: Inexact Oracle, Acceleration and Variance Reduction (NeurIPS 2024 · similarity 0.90)
- Universal Gradient Methods for Stochastic Convex Optimization (ICML 2024 · similarity 0.90)
- A Study of First-Order Methods with a Deterministic Relative-Error Gradient Oracle (ICML 2024 · similarity 0.90)
- The power of first-order smooth optimization for black-box non-smooth problems (ICML 2022 · similarity 0.89)
- A Universal Transfer Theorem for Convex Optimization Algorithms Using Inexact First-order Oracles (ICML 2024 · similarity 0.87)
- Adaptive Proximal Gradient Methods Are Universal Without Approximation (ICML 2024 · 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.