ICLR 2025 accepted papers on Bandits and online learning
38 of the 3,703 papers accepted at ICLR 2025 (1.0%) are on Bandits and online learning, compared with 1.1% at ICLR 2024. They fall into 4 topics.
See ICLR 2025 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
Contextual bandits and Thompson sampling · 13 papers
- Lasso Bandit with Compatibility Condition on Optimal Arm (1 citation)
- Bayesian Analysis of Combinatorial Gaussian Process Bandits
- Conservative Contextual Bandits: Beyond Linear Representations
- Demystifying Online Clustering of Bandits: Enhanced Exploration Under Stochastic and Smoothed Adversarial Contexts
- Dynamic Assortment Selection and Pricing with Censored Preference Feedback
- Finally Rank-Breaking Conquers MNL Bandits: Optimal and Efficient Algorithms for MNL Assortment
- Geometry-Aware Approaches for Balancing Performance and Theoretical Guarantees in Linear Bandits
- Learning from Imperfect Human Feedback: A Tale from Corruption-Robust Dueling
- Neural Dueling Bandits: Preference-Based Optimization with Human Feedback
- Second Order Bounds for Contextual Bandits with Function Approximation
- Almost Optimal Batch-Regret Tradeoff for Batch Linear Contextual Bandits
- Efficient Top-m Data Values Identification for Data Selection
and 1 more in the interactive map.
Online convex optimization and games · 11 papers
- Generalized Principal-Agent Problem with a Learning Agent (1 citation)
- Regret Bounds for Episodic Risk-Sensitive Linear Quadratic Regulator (1 citation)
- Feature-Based Online Bilateral Trade
- Last-Iterate Convergence Properties of Regret-Matching Algorithms in Games
- Learning Successor Features with Distributed Hebbian Temporal Memory
- On the Convergence of No-Regret Dynamics in Information Retrieval Games with Proportional Ranking Functions
- uniINF: Best-of-Both-Worlds Algorithm for Parameter-Free Heavy-Tailed MABs
- An Online Learning Theory of Trading-Volume Maximization
- Bandit Learning in Matching Markets with Indifference
- Competitive Fair Scheduling with Predictions
- Minimalistic Predictions for Online Class Constraint Scheduling
Multi-armed bandits · 9 papers
- Be More Diverse than the Most Diverse: Optimal Mixtures of Generative Models via Mixture-UCB Bandit Algorithms
- DOPL: Direct Online Preference Learning for Restless Bandits with Preference Feedback
- On Bits and Bandits: Quantifying the Regret-Information Trade-off
- Pairwise Elimination with Instance-Dependent Guarantees for Bandits with Cost Subsidy
- Stochastic Bandits Robust to Adversarial Attacks
- Breaking the $\log(1/\Delta_2)$ Barrier: Better Batched Best Arm Identification with Adaptive Grids
- Lipschitz Bandits in Optimal Space
- Regret-Optimal List Replicable Bandit Learning: Matching Upper and Lower Bounds
- Satisficing Regret Minimization in Bandits
RL theory and MDPs · 5 papers
- Gap-Dependent Bounds for Q-Learning using Reference-Advantage Decomposition
- Minimax Optimal Reinforcement Learning with Quasi-Optimism
- Misspecified $Q$-Learning with Sparse Linear Function Approximation: Tight Bounds on Approximation Error
- Optimal Non-Asymptotic Rates of Value Iteration for Average-Reward Markov Decision Processes
- Optimal Strong Regret and Violation in Constrained MDPs via Policy Optimization