Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Linear and generalized linear bandits
Derive improved regret bounds for stochastic linear bandits with efficient updates.
88 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Contextual bandits and Thompson sampling, keyword cluster “contextual bandits · contexts · user”
State History Tried Checklist 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.
24 papers in 2023–24 and 22 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 16.1% to 16.5%.
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
- 2016 · Distributed Clustering of Linear Bandits in Peer to Peer Networks (ICML 2016 · 96 citations)
- 2017 · Misspecified Linear Bandits (AAAI 2017 · 80 citations)
- 2019 · Bilinear Bandits with Low-rank Structure (ICML 2019 · 13 citations)
- 2020 · Meta-learning with Stochastic Linear Bandits (ICML 2020 · 23 citations)
- 2021 · An Efficient Pessimistic-Optimistic Algorithm for Stochastic Linear Bandits with General Constraints (NeurIPS 2021 · 17 citations)
- 2022 · Batch-Size Independent Regret Bounds for Combinatorial Semi-Bandits with Probabilistically Triggered Arms or Independent Arms (NeurIPS 2022 · 6 citations)
- 2023 · Combinatorial Causal Bandits (AAAI 2023 · 5 citations)
- 2024 · Hierarchize Pareto Dominance in Multi-Objective Stochastic Linear Bandits (AAAI 2024 · 2 citations)
- 2025 · Lasso Bandit with Compatibility Condition on Optimal Arm (ICLR 2025 · 1 citation)
- 2026 · Stochastic Linear Bandits with Parameter Noise (ICML 2026 · 0 citations)
Earliest papers here
- Distributed Clustering of Linear Bandits in Peer to Peer Networks (ICML 2016 · 96 citations)
- Online Stochastic Linear Optimization under One-bit Feedback (ICML 2016 · 30 citations)
- Linear Submodular Bandits with a Knapsack Constraint (AAAI 2016 · 17 citations)
What has been tried
68 papers here report 399 measured changes (a component added, removed or swapped, compared with a baseline): helped 69%, no clear effect 10%, hurt 15%, mixed 6%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Use deterministic rather than randomized action sampling | Stochastic linear bandits · regret lower bound (weak corruption C∞) | \Omega(d\sqrt{T}+dC_\infty) | Hurt | Corruption-Robust Linear Bandits: Minimax Optimality and Gap-Dependent Misspecification (NeurIPS 2024 · 1 citation) |
| Increase the threshold μ | Synthetic Example 2 and Segment · fraction of OFUL actions | Hurt | Geometry-Aware Approaches for Balancing Performance and Theoretical Guarantees in Linear Bandits (ICLR 2025 · 0 citations) | |
| Set working sparsity s=5, below the true sparsity s_0=10 | Sparse recovery · support recovery rate | Hurt | High-dimensional Linear Bandits with Knapsacks (ICML 2024 · 0 citations) | |
| Increase the instance parameter ε | End of Optimism instances · regret | Hurt | Optimal Batched Linear Bandits (ICML 2024 · 0 citations) | |
| include the ground-truth subspace among the experts | Synthetic adversarial environment · cumulative regret | Helped | Beyond task diversity: provable representation transfer for sequential multitask linear bandits (NeurIPS 2024 · 0 citations) | |
| Add the context-dependent Switching Criterion I | Generalized linear contextual bandits, M2 · regret | Helped | Generalized Linear Bandits with Limited Adaptivity (NeurIPS 2024 · 0 citations) | |
| Use the theory-based confidence-set distribution instead of a uniform distribution | Synthetic model selection; S=1,2,4,8,16 · cumulative regret | Helped | Sparsity-Agnostic Linear Bandits with Adaptive Adversaries (NeurIPS 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 15 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Baselines
- LinTS · 2 of 15 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 46 papers; from OpenAlex and the papers' affiliations).
- Seoul National University · 5 papers
- Microsoft · 4 papers
- National University of Singapore · 3 papers
- University of Arizona · 3 papers
- Tsinghua University · 2 papers
- Columbia University · 2 papers
- City University of Hong Kong · 2 papers
- The University of Tokyo · 2 papers
- RIKEN AIP · 2 papers
- University of Washington · 2 papers
Use it
Download BibTeX of 40 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
- Oracle-efficient contextual bandits (168 papers)
- Practical contextual bandit extensions (77 papers)
- Causal and structured bandits (61 papers)
- Ranking bandits (42 papers)
Elsewhere in the topic
- Thompson sampling analysis (84 papers)
- Gaussian process bandit regret (26 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).