ICLR 2017 accepted papers on Reinforcement learning
18 of the 198 papers accepted at ICLR 2017 (9.1%) are on Reinforcement learning. They fall into 3 topics.
See ICLR 2017 in the interactive map
Submitting to ICLR? 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 · 10 papers
- An Actor-Critic Algorithm for Sequence Prediction (215 citations)
- Sample Efficient Actor-Critic with Experience Replay (210 citations)
- Learning to Optimize (153 citations)
- Q-Prop: Sample-Efficient Policy Gradient with An Off-Policy Critic (117 citations)
- Combining policy gradient and Q-learning (89 citations)
- Learning to Play in a Day: Faster Deep Reinforcement Learning by Optimality Tightening (56 citations)
- Learning to Repeat: Fine Grained Action Repetition for Deep Reinforcement Learning (29 citations)
- Reinforcement Learning through Asynchronous Advantage Actor-Critic on a GPU (28 citations)
- Batch Policy Gradient Methods for Improving Neural Conversation Models (7 citations)
- Improving Policy Gradient by Exploring Under-appreciated Rewards (5 citations)
Visual and representation RL · 7 papers
- Reinforcement Learning with Unsupervised Auxiliary Tasks (259 citations)
- Stochastic Neural Networks for Hierarchical Reinforcement Learning (184 citations)
- Learning Visual Servoing with Deep Features and Fitted Q-Iteration (51 citations)
- Learning to Compose Words into Sentences with Reinforcement Learning (42 citations)
- Metacontrol for Adaptive Imagination-Based Optimization (42 citations)
- Generalizing Skills with Semi-Supervised Reinforcement Learning (34 citations)
- Episodic Exploration for Deep Deterministic Policies for StarCraft Micromanagement (7 citations)
Imitation learning · 1 paper
- Third Person Imitation Learning (146 citations)
