NeurIPS 2017 accepted papers on Reinforcement learning
36 of the 688 papers accepted at NeurIPS 2017 (5.2%) are on Reinforcement learning, compared with 3.9% at NeurIPS 2016. They fall into 5 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
Policy gradient and actor-critic · 16 papers
- Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation (461 citations)
- Safe Model-based Reinforcement Learning with Stability Guarantees (319 citations)
- Bridging the Gap Between Value and Policy Based Reinforcement Learning (214 citations)
- Hybrid Reward Architecture for Reinforcement Learning (185 citations)
- Concrete Dropout (137 citations)
- Reliable Decision Support using Counterfactual Models (86 citations)
- Off-policy evaluation for slate recommendation (39 citations)
- Shallow Updates for Deep Reinforcement Learning (24 citations)
- Using Options and Covariance Testing for Long Horizon Off-Policy Policy Evaluation (23 citations)
- Cold-Start Reinforcement Learning with Softmax Policy Gradient (21 citations)
- Policy Gradient With Value Function Approximation For Collective Multiagent Planning (14 citations)
- Reinforcement Learning under Model Mismatch (10 citations)
and 4 more in the interactive map.
Visual and representation RL · 11 papers
- Hindsight Experience Replay (337 citations)
- Imagination-Augmented Agents for Deep Reinforcement Learning (209 citations)
- Thinking Fast and Slow with Deep Learning and Tree Search (190 citations)
- Successor Features for Transfer in Reinforcement Learning (183 citations)
- Distral: Robust multitask reinforcement learning (175 citations)
- EX2: Exploration with Exemplar Models for Deep Reinforcement Learning (59 citations)
- Value Prediction Network (40 citations)
- QMDP-Net: Deep Learning for Planning under Partial Observability (27 citations)
- Towards Generalization and Simplicity in Continuous Control (26 citations)
- DPSCREEN: Dynamic Personalized Screening (6 citations)
- Dynamic-Depth Context Tree Weighting (1 citation)
MDPs and planning · 4 papers
- Finite Sample Analysis of the GTD Policy Evaluation Algorithms in Markov Setting (23 citations)
- Data-Efficient Reinforcement Learning in Continuous State-Action Gaussian-POMDPs (19 citations)
- Is the Bellman residual a bad proxy?
- #Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
Imitation learning · 4 papers
- Robust Imitation of Diverse Behaviors (114 citations)
- Inverse Reward Design (61 citations)
- State Aware Imitation Learning (20 citations)
- Compatible Reward Inverse Reinforcement Learning (17 citations)
RL from feedback · 1 paper
- Deep Reinforcement Learning from Human Preferences (500 citations)
