Research map / Bandits and online learning / RL theory and MDPs
Online Reinforcement Learning in Stochastic 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 |
|---|---|---|---|
| Use UCSG instead of R-max for online learning under the comparable average-reward SG setting | Online average-reward stochastic games · Average-reward SG, Assumption 2 | regret: R-max: \tilde{\mathcal{O}}\left(\frac{D^3S^2A}{\varepsilon^6}\right); UCSG: \tilde{\mathcal{O}}(\sqrt[3]{DS^2AT^2}) | Helped |
| Use UCSG's offline training procedure to find a maximin stationary policy | UCSG · Average-reward SG, Assumption 1 (offline) | sample complexity: \tilde{\mathcal{O}}(D^3S^5A+D^2S^2A/\varepsilon^2) | Helped |
| Use UCSG's offline training procedure under the additional reachability assumption | UCSG · Average-reward SG, Assumption 2 plus additional reachability assumption (offline) | sample complexity: \tilde{\mathcal{O}}(DS^2A/\varepsilon^3) | Helped |
Closest papers by meaning
- Learning in Markov Games with Adaptive Adversaries: Policy Regret, Fundamental Barriers, and Efficient Algorithms (NeurIPS 2024 · similarity 0.90)
- Optimism Without Regularization: Constant Regret in Zero-Sum Games (NeurIPS 2025 · similarity 0.89)
- Policy-Regret Minimization in Markov Games with Function Approximation (ICML 2025 · similarity 0.89)
- No-regret Learning in Harmonic Games: Extrapolation in the Face of Conflicting Interests (NeurIPS 2024 · similarity 0.89)
- Incentivize without Bonus: Provably Efficient Model-based Online Multi-agent RL for Markov Games (ICML 2025 · similarity 0.89)
- Best of Both Worlds: Regret Minimization versus Minimax Play (ICML 2025 · similarity 0.89)
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.