ICLR 2018 accepted papers on Reinforcement learning
36 of the 337 papers accepted at ICLR 2018 (10.7%) are on Reinforcement learning, compared with 9.1% at ICLR 2017. They fall into 4 topics.
See ICLR 2018 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
Visual and representation RL · 16 papers
- A Deep Reinforced Model for Abstractive Summarization (1,255 citations)
- Parameter Space Noise for Exploration (359 citations)
- Learning an Embedding Space for Transferable Robot Skills (175 citations)
- Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments (150 citations)
- NerveNet: Learning Structured Policy with Graph Neural Networks (145 citations)
- Emergent Complexity via Multi-Agent Competition (143 citations)
- N2N learning: Network to Network Compression via Policy Gradient Reinforcement Learning (135 citations)
- META LEARNING SHARED HIERARCHIES (111 citations)
- Leave no Trace: Learning to Reset for Safe and Autonomous Reinforcement Learning (73 citations)
- Eigenoption Discovery through the Deep Successor Representation (71 citations)
- Unsupervised Learning of Goal Spaces for Intrinsically Motivated Goal Exploration (57 citations)
- Hierarchical and Interpretable Skill Acquisition in Multi-task Reinforcement Learning (55 citations)
and 4 more in the interactive map.
Policy gradient and actor-critic · 16 papers
- Distributed Distributional Deterministic Policy Gradients (320 citations)
- Model-Ensemble Trust-Region Policy Optimization (261 citations)
- Maximum a Posteriori Policy Optimisation (225 citations)
- Temporal Difference Models: Model-Free Deep RL for Model-Based Control (152 citations)
- Variance Reduction for Policy Gradient with Action-Dependent Factorized Baselines (83 citations)
- DORA The Explorer: Directed Outreaching Reinforcement Action-Selection (36 citations)
- Action-dependent Control Variates for Policy Optimization via Stein Identity (33 citations)
- Trust-PCL: An Off-Policy Trust Region Method for Continuous Control (28 citations)
- TreeQN and ATreeC: Differentiable Tree-Structured Models for Deep Reinforcement Learning (25 citations)
- The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning (24 citations)
- Policy Optimization by Genetic Distillation (11 citations)
- Divide-and-Conquer Reinforcement Learning (10 citations)
and 4 more in the interactive map.
