NeurIPS 2017 accepted papers on Bandits and online learning
42 of the 688 papers accepted at NeurIPS 2017 (6.1%) are on Bandits and online learning, compared with 6.5% at NeurIPS 2016. They fall into 4 topics.
See NeurIPS 2017 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 · 21 papers
- Improved Dynamic Regret for Non-degenerate Functions (66 citations)
- Learning with Bandit Feedback in Potential Games (53 citations)
- Online Influence Maximization under Independent Cascade Model with Semi-Bandit Feedback (52 citations)
- Online Convex Optimization with Stochastic Constraints (28 citations)
- Online Learning for Multivariate Hawkes Processes (26 citations)
- Parameter-Free Online Learning via Model Selection (23 citations)
- Online Learning with a Hint (12 citations)
- Stochastic and Adversarial Online Learning without Hyperparameters (9 citations)
- Near Minimax Optimal Players for the Finite-Time 3-Expert Prediction Problem (8 citations)
- Regret Analysis for Continuous Dueling Bandit (7 citations)
- Online multiclass boosting (6 citations)
- Multi-Armed Bandits with Metric Movement Costs (5 citations)
and 9 more in the interactive map.
Multi-armed bandits · 9 papers
- Minimal Exploration in Structured Stochastic Bandits (79 citations)
- Improving Regret Bounds for Combinatorial Semi-Bandits with Probabilistically Triggered Arms and Its Applications (31 citations)
- Rotting Bandits (30 citations)
- Boltzmann Exploration Done Right (25 citations)
- Identifying Outlier Arms in Multi-Armed Bandit (14 citations)
- Position-based Multiple-play Bandit Problem with Unknown Position Bias (13 citations)
- Bandits Dueling on Partially Ordered Sets (1 citation)
- A KL-LUCB algorithm for Large-Scale Crowdsourcing
- A Scale Free Algorithm for Stochastic Bandits with Bounded Kurtosis
Contextual bandits and Thompson sampling · 6 papers
- Scalable Generalized Linear Bandits: Online Computation and Hashing (60 citations)
- Multi-Task Learning for Contextual Bandits (50 citations)
- Conservative Contextual Linear Bandits (38 citations)
- Interactive Submodular Bandit (14 citations)
- Real-Time Bidding with Side Information (8 citations)
- Action Centered Contextual Bandits
RL theory and MDPs · 6 papers
- Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning (138 citations)
- Learning Unknown Markov Decision Processes: A Thompson Sampling Approach (54 citations)
- Regret Minimization in MDPs with Options without Prior Knowledge (9 citations)
- Online Reinforcement Learning in Stochastic Games (7 citations)
- Posterior sampling for reinforcement learning: worst-case regret bounds (2 citations)
- Optimistic posterior sampling for reinforcement learning: worst-case regret bounds
