Research map / Bandits and online learning / Online convex optimization and games
Online Learning with Abstention
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 UCB-GT with feedback graphs estimated from past observations instead of UCB-NT with the subset feedback graph | UCB-GT · Ten datasets; abstention costs c ∈ {0.1, 0.2, 0.3} | regret | Helped |
| Share feedback across experts using the subset feedback graph rather than no feedback sharing | UCB-NT · Ten datasets | regret | Helped |
| Use a feedback graph that depends on the current input rather than one based only on past information | UCB-NT · Stochastic online learning with abstention | regret | Hurt |
| Estimate richer feedback graphs from past observations while controlling the bias of added edges | UCB-GT · Ten datasets | regret | Helped |
| Increase the abstention cost (ablation) | Online learning with abstention · Ten datasets | fraction of points abstained on | Hurt |
| Change the number of experts, expert choice, or use extreme abstention costs (ablation) | UCB-GT versus UCB-NT · Additional experiments across datasets | regret | Helped |
Closest papers by meaning
- Discounted Adaptive Online Learning: Towards Better Regularization (ICML 2024 · similarity 0.89)
- A Simple and Optimal Approach for Universal Online Learning with Gradient Variations (NeurIPS 2024 · similarity 0.89)
- No-Regret Learning Under Adversarial Resource Constraints: A Spending Plan Is All You Need! (NeurIPS 2025 · similarity 0.88)
- High-Dimensional Prediction for Sequential Decision Making (ICML 2025 · similarity 0.88)
- Bandits with Abstention under Expert Advice (NeurIPS 2024 · similarity 0.88)
- Online Learning in the Random-Order Model (ICML 2025 · 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.