ICML 2026 accepted papers on Bandits and online learning
88 of the 6,341 papers accepted at ICML 2026 (1.4%) are on Bandits and online learning, compared with 2.9% at ICML 2025. They fall into 4 topics.
See ICML 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Online convex optimization and games · 43 papers
- A Perturbation Approach to Unconstrained Linear Bandits
- A Stronger Benchmark for Online Bilateral Trade: From Fixed Prices to Distributions
- A Two-Layer Framework for Joint Online Configuration Selection and Admission Control
- Achieving Logarithmic Regret in KL-Regularized Zero-Sum Markov Games
- DAL: A Practical Prior-Free Black-Box Framework for Piecewise Stationary Bandits
- Dynamic Regret via Discounted-to-Dynamic Reduction with Applications to Curved Losses and Adam Optimizer
- Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets
- Follow-the-Perturbed-Leader for Decoupled Bandits: Best-of-Both-Worlds and Practicality
- Generalized Linear Bandits with Memory
- Learning in Bayesian Stackelberg Games With Unknown Follower's Types
- Learning in Structured Stackelberg Games
- Learning-Augmented Online Covering Problems
and 31 more in the interactive map.
Contextual bandits and Thompson sampling · 22 papers
- Contextual Rollout Bandits for Reinforcement Learning with Verifiable Rewards (1 citation)
- A Direct Approach for Handling Contextual Bandits with Latent State Dynamics
- Adaptive Bandit Algorithms for Contextual Matching Markets
- BFTS: Thompson Sampling with Bayesian Additive Regression Trees
- Contextual Slate GLM Bandits with Limited Adaptivity
- Incentivized Exploration with Stochastic Covariates: A Two-Stage Mechanism Design for Recommender System
- Keep Everyone Happy: Online Fair Division of Numerous Items with Few Copies
- Linear Bandits beyond Inner Product Spaces, the case of Bandit Optimal Transport
- Neural Logistic Bandits
- On Regret Bounds of Thompson Sampling for Bayesian Optimization
- Optimal Regret for Policy Optimization in Contextual Bandits
- Practical and Optimal Algorithm for Linear Contextual Bandits with Rare Parameter Updates
and 10 more in the interactive map.
Multi-armed bandits · 14 papers
- Box Thirding: Anytime Best Arm Identification under Insufficient Sampling
- Cutting LLM Evaluation Costs with SySRs: A Bandit Algorithm That Provably Exploits Model Similarity
- Fixed Budget is No Harder Than Fixed Confidence in Best-Arm Identification up to Logarithmic Factors
- Improved Algorithms for Nash Welfare in Linear Bandits
- KernelBand: Steering LLM-based Kernel Optimization via Hardware-Aware Multi-Armed Bandits
- Near-Optimal Regret for KL-Regularized Multi-Armed Bandits
- Online Learning with Recency: Algorithms for Sliding-window Streaming Multi-armed Bandits
- Optimal Top-$k$ Identification from Pairwise Comparisons
- Variance Driven Exploration: A Provable and Efficient Methodology for Pure Exploration in Highly Stochastic Environments
- Bandit Social Leaning Dynamics with Exploration Episodes
- Data-Source Adaptive Online Learning under Heteroscedastic Noise
- Exploration-free Algorithms for Multi-group Mean Estimation
and 2 more in the interactive map.
RL theory and MDPs · 9 papers
- Best-of-Both-Worlds for Heavy-Tailed Markov Decision Processes
- Data- and Variance-dependent Regret Bounds for Online Tabular MDPs
- Exact Unlearning in Reinforcement Learning
- Minimax Optimal Strategy for Delayed Observations in Online Reinforcement Learning
- Near-Optimal Regret for Policy Optimization in Contextual MDPs with General Offline Function Approximation
- Posterior Sampling Reinforcement Learning with Gaussian Processes for Continuous Control: Sublinear Regret Bounds for Unbounded State Spaces
- Near-Minimax Multi-Objective RL under Predictable Adversarial Preferences and Preference-Free Exploration in Linear MDPs
- Rethinking the Hardness of PbRL: A Provable General Regret Bound
- Towards Achieving Optimal Strong Regret and Constraint Violation via Computationally Efficient Model-free RL