Research map / Bandits and online learning / Online convex optimization and games
Learning with Bandit Feedback in Potential Games
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 |
|---|---|---|---|
| Add exploration factor epsilon>0 (action probabilities bounded below by epsilon) to exponential weights under bandit feedback | Exponential weights (epsilon-Hedge) with bandit estimator · Potential games, bandit feedback | convergence to delta-equilibrium (delta -> 0 as epsilon -> 0) | Helped |
| Decreasing exploration factor eps_n -> 0 with step-size conditions (gamma_n/eps_n^2 -> 0, summable gamma_n^2/eps_n, etc.) | Exponential weights with annealing (Algorithm 2) under bandit feedback · Potential games, bandit feedback | convergence to exact Nash equilibrium | Helped |
| Use step-size exponent beta in (1/2,1] with annealed exploration; beta=1/2 precludes any vanishing eps_n satisfying condition (4.6) | Algorithm 2 (annealed bandit EW) · Potential games, bandit feedback | feasibility of convergence conditions | Hurt |
| Semi-bandit feedback (noisy estimates of full payoff vector) instead of bandit feedback, with zero exploration | Exponential weights (Hedge) · Potential games, semi-bandit feedback | convergence to Nash equilibrium (a.s.) | Helped |
Closest papers by meaning
- The Harder Path: Last Iterate Convergence for Uncoupled Learning in Zero-Sum Games with Bandit Feedback (ICML 2025 · similarity 0.92)
- Near-Optimal Last-Iterate Convergence for Zero-Sum Games with Bandit Feedback and Opponent Actions (NeurIPS 2026 · similarity 0.91)
- Uncoupled and Convergent Learning in Monotone Games under Bandit Feedback (NeurIPS 2025 · similarity 0.91)
- Learning Imperfect Information Extensive-form Games with Last-iterate Convergence under Bandit Feedback (ICML 2025 · similarity 0.91)
- Convergence of No-Swap-Regret Dynamics in Self-Play (NeurIPS 2024 · similarity 0.90)
- No-regret Learning in Harmonic Games: Extrapolation in the Face of Conflicting Interests (NeurIPS 2024 · similarity 0.90)
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.