NeurIPS 2018 accepted papers on Reinforcement learning
70 of the 1,008 papers accepted at NeurIPS 2018 (6.9%) are on Reinforcement learning, compared with 5.2% at NeurIPS 2017. They fall into 4 topics.
See NeurIPS 2018 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
Policy gradient and actor-critic · 30 papers
- Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models (180 citations)
- Meta-Gradient Reinforcement Learning (177 citations)
- Simple random search of static linear policies is competitive for reinforcement learning (144 citations)
- Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansion (143 citations)
- Evolved Policy Gradients (133 citations)
- Randomized Prior Functions for Deep Reinforcement Learning (104 citations)
- Memory Augmented Policy Optimization for Program Synthesis and Semantic Parsing (96 citations)
- Breaking the Curse of Horizon: Infinite-Horizon Off-Policy Estimation (94 citations)
- Deep Reinforcement Learning of Marked Temporal Point Processes (70 citations)
- Balanced Policy Evaluation and Learning (69 citations)
- Actor-Critic Policy Optimization in Partially Observable Multiagent Environments (67 citations)
- Evolution-Guided Policy Gradient in Reinforcement Learning (66 citations)
and 18 more in the interactive map.
Visual and representation RL · 22 papers
- Recurrent World Models Facilitate Policy Evolution (452 citations)
- Visual Reinforcement Learning with Imagined Goals (307 citations)
- Data-Efficient Hierarchical Reinforcement Learning (248 citations)
- Differentiable MPC for End-to-end Planning and Control (198 citations)
- Meta-Reinforcement Learning of Structured Exploration Strategies (174 citations)
- Reinforcement Learning for Solving the Vehicle Routing Problem (169 citations)
- Verifiable Reinforcement Learning via Policy Extraction (159 citations)
- Improving Exploration in Evolution Strategies for Deep Reinforcement Learning via a Population of Novelty-Seeking Agents (144 citations)
- Learn What Not to Learn: Action Elimination with Deep Reinforcement Learning (125 citations)
- Data center cooling using model-predictive control (114 citations)
- Diversity-Driven Exploration Strategy for Deep Reinforcement Learning (69 citations)
- M-Walk: Learning to Walk over Graphs using Monte Carlo Tree Search (58 citations)
and 10 more in the interactive map.
Imitation learning · 9 papers
- Variational Inverse Control with Events: A General Framework for Data-Driven Reward Definition (53 citations)
- Exponentially Weighted Imitation Learning for Batched Historical Data (45 citations)
- Learning Task Specifications from Demonstrations (39 citations)
- Reward learning from human preferences and demonstrations in Atari (39 citations)
- Teaching Inverse Reinforcement Learners via Features and Demonstrations (36 citations)
- Where Do You Think You're Going?: Inferring Beliefs about Dynamics from Behavior (26 citations)
- A Bayesian Approach to Generative Adversarial Imitation Learning (15 citations)
- Learning Safe Policies with Expert Guidance (6 citations)
- Lifelong Inverse Reinforcement Learning
MDPs and planning · 9 papers
- Is Q-Learning Provably Efficient? (306 citations)
- A Lyapunov-based Approach to Safe Reinforcement Learning (77 citations)
- Dual Policy Iteration (19 citations)
- On Oracle-Efficient PAC RL with Rich Observations (17 citations)
- Policy-Conditioned Uncertainty Sets for Robust Markov Decision Processes (9 citations)
- Representation Balancing MDPs for Off-policy Policy Evaluation (8 citations)
- rho-POMDPs have Lipschitz-Continuous epsilon-Optimal Value Functions (7 citations)
- Negotiable Reinforcement Learning for Pareto Optimal Sequential Decision-Making (4 citations)
- Temporal Regularization for Markov Decision Process
