NeurIPS 2024 accepted papers on Bandits and online learning
128 of the 4,560 papers accepted at NeurIPS 2024 (2.8%) are on Bandits and online learning, compared with 3.6% at NeurIPS 2023. They fall into 4 topics.
See NeurIPS 2024 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 · 60 papers
- Optimal Algorithms for Online Convex Optimization with Adversarial Constraints (7 citations)
- Improved Algorithms for Contextual Dynamic Pricing (2 citations)
- Learning-Augmented Algorithms for the Bahncard Problem (2 citations)
- Learning-Augmented Priority Queues (2 citations)
- MAC Advice for facility location mechanism design (2 citations)
- Overcoming Brittleness in Pareto-Optimal Learning Augmented Algorithms (2 citations)
- A Best-of-both-worlds Algorithm for Bandits with Delayed Feedback with Robustness to Excessive Delays (1 citation)
- Assouad, Fano, and Le Cam with Interaction: A Unifying Lower Bound Framework and Characterization for Bandit Learnability (1 citation)
- Beyond Primal-Dual Methods in Bandits with Stochastic and Adversarial Constraints (1 citation)
- Convergence of No-Swap-Regret Dynamics in Self-Play (1 citation)
- Efficient $\Phi$-Regret Minimization with Low-Degree Swap Deviations in Extensive-Form Games (1 citation)
- Fully Unconstrained Online Learning (1 citation)
and 48 more in the interactive map.
Contextual bandits and Thompson sampling · 37 papers
- Causal Bandits for Linear Structural Equation Models (5 citations)
- Direct Preference-Based Evolutionary Multi-Objective Optimization with Dueling Bandits (4 citations)
- Optimal Scalarizations for Sublinear Hypervolume Regret (2 citations)
- Robust Neural Contextual Bandit against Adversarial Corruptions (2 citations)
- Contextual Decision-Making with Knapsacks Beyond the Worst Case (1 citation)
- Corruption-Robust Linear Bandits: Minimax Optimality and Gap-Dependent Misspecification (1 citation)
- How Does Variance Shape the Regret in Contextual Bandits? (1 citation)
- Improved Bayes Regret Bounds for Multi-Task Hierarchical Bayesian Bandit Algorithms (1 citation)
- Local Anti-Concentration Class: Logarithmic Regret for Greedy Linear Contextual Bandit (1 citation)
- Partial Structure Discovery is Sufficient for No-regret Learning in Causal Bandits (1 citation)
- RoME: A Robust Mixed-Effects Bandit Algorithm for Optimizing Mobile Health Interventions (1 citation)
- Sequential Probability Assignment with Contexts: Minimax Regret, Contextual Shtarkov Sums, and Contextual Normalized Maximum Likelihood (1 citation)
and 25 more in the interactive map.
Multi-armed bandits · 20 papers
- Exploration, Exploitation, and Engagement in Multi-Armed Bandits with Abandonment (8 citations)
- Global Rewards in Restless Multi-Armed Bandits (2 citations)
- On Weak Regret Analysis for Dueling Bandits (2 citations)
- Reciprocal Learning (2 citations)
- Locally Private and Robust Multi-Armed Bandits (1 citation)
- Optimal Multi-Fidelity Best-Arm Identification (1 citation)
- Adaptive Experimentation When You Can't Experiment
- Almost Minimax Optimal Best Arm Identification in Piecewise Stationary Linear Bandits
- An Adaptive Approach for Infinitely Many-armed Bandits under Generalized Rotting Constraints
- Bandits with Ranking Feedback
- Enhancing Preference-based Linear Bandits via Human Response Time
- Fixed Confidence Best Arm Identification in the Bayesian Setting
and 8 more in the interactive map.
RL theory and MDPs · 11 papers
- Achieving $\tilde{O}(1/\epsilon)$ Sample Complexity for Constrained Markov Decision Process (2 citations)
- Achieving Tractable Minimax Optimal Regret in Average Reward MDPs (2 citations)
- Near-Optimal Dynamic Regret for Adversarial Linear Mixture MDPs (1 citation)
- Span-Based Optimal Sample Complexity for Weakly Communicating and General Average Reward MDPs (1 citation)
- Achieving Constant Regret in Linear Markov Decision Processes
- Beating Adversarial Low-Rank MDPs with Unknown Transition and Bandit Feedback
- Local Linearity: the Key for No-regret Reinforcement Learning in Continuous MDPs
- Provably Efficient Reinforcement Learning with Multinomial Logit Function Approximation
- Randomized Exploration for Reinforcement Learning with Multinomial Logistic Function Approximation
- Uniform Last-Iterate Guarantee for Bandits and Reinforcement Learning
- Warm-up Free Policy Optimization: Improved Regret in Linear Markov Decision Processes