ICML 2020 accepted papers on Bandits and online learning
70 of the 1,087 papers accepted at ICML 2020 (6.4%) are on Bandits and online learning, compared with 5.3% at ICML 2019. They fall into 4 topics.
See ICML 2020 in the interactive map
Submitting to ICML? 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 · 26 papers
- Logarithmic Regret for Adversarial Online Control (29 citations)
- Naive Exploration is Optimal for Online LQR (27 citations)
- A new regret analysis for Adam-type algorithms (22 citations)
- Dual Mirror Descent for Online Allocation Problems (21 citations)
- Online mirror descent and dual averaging: keeping pace in the dynamic case (18 citations)
- Online Learning with Imperfect Hints (15 citations)
- Online Pricing with Offline Data: Phase Transition and Inverse Square Law (15 citations)
- Gradient-free Online Learning in Continuous Games with Delayed Rewards (13 citations)
- Parameter-free, Dynamic, and Strongly-Adaptive Online Learning (11 citations)
- Logarithmic Regret for Learning Linear Quadratic Regulators Efficiently (10 citations)
- Efficient Optimistic Exploration in Linear-Quadratic Regulators via Lagrangian Relaxation (8 citations)
- Reserve Pricing in Repeated Second-Price Auctions with Strategic Bidders (7 citations)
and 14 more in the interactive map.
Multi-armed bandits · 18 papers
- Concentration bounds for CVaR estimation: The cases of light-tailed and heavy-tailed distributions (26 citations)
- Gamification of Pure Exploration for Linear Bandits (25 citations)
- Stochastic bandits with arm-dependent delays (11 citations)
- Structure Adaptive Algorithms for Stochastic Bandits (9 citations)
- Combinatorial Pure Exploration for Dueling Bandit (7 citations)
- Best Arm Identification for Cascading Bandits in the Fixed Confidence Setting (5 citations)
- Cooperative Multi-Agent Bandits with Heavy Tails (5 citations)
- Exploration Through Reward Biasing: Reward-Biased Maximum Likelihood Estimation for Stochastic Multi-Armed Bandits (5 citations)
- The Intrinsic Robustness of Stochastic Bandits to Strategic Manipulation (5 citations)
- Bandits with Adversarial Scaling (3 citations)
- My Fair Bandit: Distributed Learning of Max-Min Fairness with Multi-player Bandits (3 citations)
- On Conditional Versus Marginal Bias in Multi-Armed Bandits (2 citations)
and 6 more in the interactive map.
Contextual bandits and Thompson sampling · 14 papers
- Neural Contextual Bandits with UCB-based Exploration (54 citations)
- Beyond UCB: Optimal and Efficient Contextual Bandits with Regression Oracles (35 citations)
- Meta-learning with Stochastic Linear Bandits (23 citations)
- Linear bandits with Stochastic Delayed Feedback (19 citations)
- Improved Optimistic Algorithms for Logistic Bandits (18 citations)
- Doubly robust off-policy evaluation with shrinkage (17 citations)
- Designing Optimal Dynamic Treatment Regimes: A Causal Reinforcement Learning Approach (14 citations)
- Thompson Sampling Algorithms for Mean-Variance Bandits (14 citations)
- Near-linear time Gaussian process optimization with adaptive batching and resparsification (6 citations)
- Structured Linear Contextual Bandits: A Sharp and Geometric Smoothed Analysis (5 citations)
- Graphical Models Meet Bandits: A Variational Thompson Sampling Approach (4 citations)
- On Approximate Thompson Sampling with Langevin Algorithms (4 citations)
and 2 more in the interactive map.
RL theory and MDPs · 12 papers
- Model-Based Reinforcement Learning with Value-Targeted Regression (63 citations)
- Learning Near Optimal Policies with Low Inherent Bellman Error (61 citations)
- Learning with Good Feature Representations in Bandits and in RL with a Generative Model (50 citations)
- Provable Self-Play Algorithms for Competitive Reinforcement Learning (44 citations)
- Reinforcement Learning in Feature Space: Matrix Bandit, Kernels, and Regret Bound (37 citations)
- Reinforcement Learning for Non-Stationary Markov Decision Processes: The Blessing of (More) Optimism (29 citations)
- Learning Adversarial Markov Decision Processes with Bandit Feedback and Unknown Transition (26 citations)
- Model-free Reinforcement Learning in Infinite-horizon Average-reward Markov Decision Processes (26 citations)
- Near-optimal Regret Bounds for Stochastic Shortest Path (17 citations)
- No-Regret Exploration in Goal-Oriented Reinforcement Learning (13 citations)
- Optimistic Policy Optimization with Bandit Feedback (10 citations)
- Tightening Exploration in Upper Confidence Reinforcement Learning (7 citations)
