NeurIPS 2021 accepted papers on Bandits and online learning
139 of the 2,508 papers accepted at NeurIPS 2021 (5.5%) are on Bandits and online learning, compared with 5.1% at NeurIPS 2020. They fall into 4 topics.
See NeurIPS 2021 in the interactive map
Submitting to NeurIPS? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Online convex optimization and games · 47 papers
- Complexity Lower Bounds for Nonconvex-Strongly-Concave Min-Max Optimization (10 citations)
- Perturbation-based Regret Analysis of Predictive Control in Linear Time Varying Systems (10 citations)
- Near-Optimal No-Regret Learning in General Games (8 citations)
- Decentralized Learning in Online Queuing Systems (6 citations)
- Dual Adaptivity: A Universal Algorithm for Minimizing the Adaptive Regret of Convex Functions (6 citations)
- Online Knapsack with Frequency Predictions (6 citations)
- Asynchronous Decentralized Online Learning (5 citations)
- Faster Matchings via Learned Duals (5 citations)
- One More Step Towards Reality: Cooperative Bandits with Imperfect Communication (5 citations)
- Last-iterate Convergence in Extensive-Form Games (4 citations)
- Learning in two-player zero-sum partially observable Markov games with perfect recall (4 citations)
- Learning-Augmented Dynamic Power Management with Multiple States via New Ski Rental Bounds (4 citations)
and 35 more in the interactive map.
Contextual bandits and Thompson sampling · 35 papers
- An Efficient Pessimistic-Optimistic Algorithm for Stochastic Linear Bandits with General Constraints (17 citations)
- Statistical Inference with M-Estimators on Adaptively Collected Data (16 citations)
- Federated Linear Contextual Bandits (15 citations)
- Post-Contextual-Bandit Inference (12 citations)
- Off-Policy Risk Assessment in Contextual Bandits (11 citations)
- Bayesian decision-making under misspecified priors with applications to meta-learning (9 citations)
- Efficient First-Order Contextual Bandits: Prediction, Allocation, and Triangular Discrimination (8 citations)
- A Domain-Shrinking based Bayesian Optimization Algorithm with Order-Optimal Regret Performance (6 citations)
- Batched Thompson Sampling (6 citations)
- Causal Bandits with Unknown Graph Structure (6 citations)
- Generalized Linear Bandits with Local Differential Privacy (6 citations)
- Optimal Order Simple Regret for Gaussian Process Bandits (5 citations)
and 23 more in the interactive map.
Multi-armed bandits · 34 papers
- Fair Algorithms for Multi-Agent Multi-Armed Bandits (16 citations)
- Online Multi-Armed Bandits with Adaptive Inference (13 citations)
- Fair Exploration via Axiomatic Bargaining (9 citations)
- Rebounding Bandits for Modeling Satiation Effects (9 citations)
- Cooperative Stochastic Bandits with Asynchronous Agents and Constrained Feedback (8 citations)
- Metadata-based Multi-Task Bandits with Bayesian Hierarchical Models (7 citations)
- Optimal Best-Arm Identification Methods for Tail-Risk Measures (7 citations)
- A Closer Look at the Worst-case Behavior of Multi-armed Bandit Algorithms (6 citations)
- Heterogeneous Multi-player Multi-armed Bandits: Closing the Gap and Generalization (6 citations)
- Differentially Private Multi-Armed Bandits in the Shuffle Model (5 citations)
- A Biased Graph Neural Network Sampler with Near-Optimal Regret (3 citations)
- Bandit Learning with Delayed Impact of Actions (3 citations)
and 22 more in the interactive map.
RL theory and MDPs · 23 papers
- Learning Policies with Zero or Bounded Constraint Violation for Constrained MDPs (16 citations)
- Regime Switching Bandits (12 citations)
- Breaking the Sample Complexity Barrier to Regret-Optimal Model-Free Reinforcement Learning (11 citations)
- Exponential Bellman Equation and Improved Regret Bounds for Risk-Sensitive Reinforcement Learning (8 citations)
- Nearly Minimax Optimal Reinforcement Learning for Discounted MDPs (8 citations)
- Online learning in MDPs with linear function approximation and bandit feedback. (8 citations)
- RL for Latent MDPs: Regret Guarantees and a Lower Bound (7 citations)
- Accommodating Picky Customers: Regret Bound and Exploration Complexity for Multi-Objective Reinforcement Learning (6 citations)
- Beyond Value-Function Gaps: Improved Instance-Dependent Regret Bounds for Episodic Reinforcement Learning (5 citations)
- Implicit Finite-Horizon Approximation and Efficient Optimal Algorithms for Stochastic Shortest Path (5 citations)
- Minimax Regret for Stochastic Shortest Path (5 citations)
- Oracle-Efficient Regret Minimization in Factored MDPs with Unknown Structure (5 citations)
and 11 more in the interactive map.
