Research map / Bandits and online learning / Online convex optimization and games
Online Learning for Structured Loss Spaces
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 |
|---|---|---|---|
| Adapt the OMD regularizer to s-sparse losses | OMD for prediction with expert advice · s-sparse losses | regret upper bound: 2\sqrt{{\ln (s+1)T}}; standard losses: \sqrt{{2T \ln N}} | Helped |
| Use a regularizer adapted to the low-rank loss subspace | OMD for prediction with expert advice · d-low-rank losses | regret upper bound: 4\sqrt{{dT}}; standard losses: \sqrt{{2T \ln N}} | Helped |
| Add component-specific regularizers and use their sum | OMD for additive loss spaces · d-low-rank + epsilon-noise losses | regret upper bound: \sqrt{2(16d+\epsilon)T} | Helped |
| Add component-specific regularizers and use their sum | OMD for additive loss spaces · s-sparse + epsilon-noise losses | regret upper bound: 2\sqrt{2(1+\epsilon)\ln (s+1)T} | Helped |
| Add component-specific regularizers and use their sum | OMD for additive loss spaces · d-low-rank + s-sparse losses | regret upper bound: 2\sqrt{2(16d+1)\ln (s+1)T} | Helped |
Closest papers by meaning
- Online mirror descent and dual averaging: keeping pace in the dynamic case (ICML 2020 · similarity 0.91)
- From Average Sensitivity to Small-Loss Regret Bounds under Random-Order Model (NeurIPS 2026 · similarity 0.89)
- Bandit and Delayed Feedback in Online Structured Prediction (NeurIPS 2025 · similarity 0.88)
- Online Composite Optimization Between Stochastic and Adversarial Environments (NeurIPS 2024 · similarity 0.88)
- Revisiting Projection-Free Online Learning with Time-Varying Constraints (AAAI 2025 · similarity 0.87)
- Dynamic Regret Reduces to Kernelized Static Regret (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.