Research map / Bandits and online learning / RL theory and MDPs
Average-reward and latent MDPs
Derive regret guarantees for constrained, average-reward, factored, and latent MDPs.
57 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in RL theory and MDPs, keyword cluster “markov decision · decision processes · processes mdps”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
19 papers in 2023–24 and 12 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 26.0% to 20.0%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2017 · Regret Minimization in MDPs with Options without Prior Knowledge (NeurIPS 2017 · 9 citations)
- 2018 · Efficient Bias-Span-Constrained Exploration-Exploitation in Reinforcement Learning (ICML 2018 · 4 citations)
- 2019 · Exploration Bonus for Regret Minimization in Discrete and Continuous Average Reward MDPs (NeurIPS 2019 · 20 citations)
- 2020 · Model-free Reinforcement Learning in Infinite-horizon Average-reward Markov Decision Processes (ICML 2020 · 26 citations)
- 2021 · Learning Policies with Zero or Bounded Constraint Violation for Constrained MDPs (NeurIPS 2021 · 16 citations)
- 2022 · Learning Infinite-horizon Average-reward Markov Decision Process with Constraints (ICML 2022 · 3 citations)
- 2023 · Regret Bounds for Markov Decision Processes with Recursive Optimized Certainty Equivalents (ICML 2023 · 5 citations)
- 2024 · Regret Analysis of Policy Gradient Algorithm for Infinite Horizon Average Reward Markov Decision Processes (AAAI 2024 · 2 citations)
- 2025 · Quantum Speedups in Regret Analysis of Infinite Horizon Average-Reward Markov Decision Processes (ICML 2025 · 1 citation)
- 2026 · Minimax Optimal Strategy for Delayed Observations in Online Reinforcement Learning (ICML 2026 · 0 citations)
Earliest papers here
- Regret Minimization in MDPs with Options without Prior Knowledge (NeurIPS 2017 · 9 citations)
- Efficient Bias-Span-Constrained Exploration-Exploitation in Reinforcement Learning (ICML 2018 · 4 citations)
- Exploration Bonus for Regret Minimization in Discrete and Continuous Average Reward MDPs (NeurIPS 2019 · 20 citations)
What has been tried
45 papers here report 190 measured changes (a component added, removed or swapped, compared with a baseline): helped 75%, no clear effect 9%, hurt 13%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Use Legendre polynomial features instead of standard polynomial features | Modified LQR, both environments · episodic return | Helped | No-Regret Reinforcement Learning in Smooth MDPs (ICML 2024 · 1 citation) | |
| Replace Hoeffding bonuses with Bernstein bonuses | Gap-dependent federated tabular MDPs · regret | No clear effect | Gap-Dependent Bounds for Federated $Q$-Learning (ICML 2025 · 0 citations) | |
| Use RBF instead of Matérn kernels | GP-sampled synthetic MDPs · cumulative regret | Helped | No-Regret Thompson Sampling for Finite-Horizon Markov Decision Processes with Gaussian Processes (NeurIPS 2025 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 31 papers; from OpenAlex and the papers' affiliations).
- Princeton University · 4 papers
- University of Washington · 3 papers
- Meta · 3 papers
- Purdue University · 2 papers
- University of Pennsylvania · 2 papers
- Google · 2 papers
- Amazon · 2 papers
- Google DeepMind · 2 papers
- University of California, Los Angeles · 2 papers
Use it
Download BibTeX of 39 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Non-stationary and structured MDPs (30 papers)
- Adversarial MDPs with bandit feedback (30 papers)
Elsewhere in the topic
- Regret bounds with function approximation (79 papers)
- Linear function approximation in RL (64 papers)
- Stochastic shortest path (11 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).