Research map / Bandits and online learning / Online convex optimization and games
Second-Order Kernel Online Convex Optimization with Adaptive Sketching
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 |
|---|---|---|---|
| Exploit second-order curvature with kernelized online Newton updates | KOCO · Exp-concave losses | regret upper bound: \mathcal{O}(\sqrt{T}) -> \mathcal{O}( f^t\log T) | Helped |
| Use an adaptive low-rank sketch controlled by γ | KONS · Exp-concave KOCO | time and space per iteration: \mathcal{O}(t^2) -> \mathcal{O}(t^2\gamma^2) | Helped |
| Use an adaptive low-rank sketch controlled by γ | KONS · Exp-concave KOCO | regret upper bound: 1/γ times more regret | Hurt |
| Set a positive lower bound γ on sampling probabilities | Sketched-KONS · Exp-concave KOCO | regret upper bound: 1/γ increase in regret | Hurt |
| Set a positive lower bound γ on sampling probabilities | Sketched-KONS · Exp-concave KOCO | time and space per iteration: γ² decrease | Helped |
| Use curvature-aware second-order updates rather than first-order GD | KOCO with exp-concave losses · Exp-concave losses | regret upper bound: \mathcal{O}(\sqrt{T}) -> \mathcal{O}( f^T(\alpha)\log T) | Helped |
Closest papers by meaning
- Dynamic Regret Reduces to Kernelized Static Regret (NeurIPS 2025 · similarity 0.87)
- Fast Rates in Stochastic Online Convex Optimization by Exploiting the Curvature of Feasible Sets (NeurIPS 2024 · similarity 0.87)
- From Non-Convex to Strongly Convex: Curvature-Adaptive FTPL for Online Optimization (NeurIPS 2026 · similarity 0.87)
- Exploiting Curvature in Online Convex Optimization with Delayed Feedback (ICML 2025 · similarity 0.86)
- Universal Online Convex Optimization with $1$ Projection per Round (NeurIPS 2024 · similarity 0.86)
- Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization (NeurIPS 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.