ICLR 2020 accepted papers on Reinforcement learning
81 of the 687 papers accepted at ICLR 2020 (11.8%) are on Reinforcement learning, compared with 12.5% at ICLR 2019. They fall into 5 topics.
See ICLR 2020 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 · 43 papers
- Model Based Reinforcement Learning for Atari (401 citations)
- Causal Discovery with Reinforcement Learning (109 citations)
- VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning (94 citations)
- Adversarial Policies: Attacking Deep Reinforcement Learning (90 citations)
- Network Randomization: A Simple Technique for Generalization in Deep Reinforcement Learning (81 citations)
- Exploring Model-based Planning with Policy Networks (80 citations)
- Never Give Up: Learning Directed Exploration Strategies (72 citations)
- Dynamics-Aware Unsupervised Discovery of Skills (70 citations)
- Improving Generalization in Meta Reinforcement Learning using Learned Objectives (54 citations)
- Measuring the Reliability of Reinforcement Learning Algorithms (52 citations)
- Model-based reinforcement learning for biological sequence design (46 citations)
- Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learning (44 citations)
and 31 more in the interactive map.
Policy gradient and actor-critic · 22 papers
- Implementation Matters in Deep RL: A Case Study on PPO and TRPO (116 citations)
- Projection-Based Constrained Policy Optimization (77 citations)
- ES-MAML: Simple Hessian-Free Meta Learning (75 citations)
- V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control (51 citations)
- Model-Augmented Actor-Critic: Backpropagating through Paths (42 citations)
- Robust Reinforcement Learning for Continuous Control with Model Misspecification (41 citations)
- Sample Efficient Policy Gradient Methods with Recursive Variance Reduction (36 citations)
- Doubly Robust Bias Reduction in Infinite Horizon Off-Policy Estimation (29 citations)
- Learning to solve the credit assignment problem (28 citations)
- A Closer Look at Deep Policy Gradients (26 citations)
- Harnessing Structures for Value-Based Planning and Reinforcement Learning (15 citations)
- CAQL: Continuous Action Q-Learning (12 citations)
and 10 more in the interactive map.
Imitation learning · 10 papers
- Deep Imitative Models for Flexible Inference, Planning, and Control (56 citations)
- SQIL: Imitation Learning via Reinforcement Learning with Sparse Rewards (51 citations)
- Disagreement-Regularized Imitation Learning (40 citations)
- State Alignment-based Imitation Learning (27 citations)
- Imitation Learning via Off-Policy Distribution Matching (26 citations)
- Maximum Likelihood Constraint Inference for Inverse Reinforcement Learning (22 citations)
- On Computation and Generalization of Generative Adversarial Imitation Learning (9 citations)
- State-only Imitation with Transition Dynamics Mismatch (8 citations)
- Watch, Try, Learn: Meta-Learning from Demonstrations and Rewards (8 citations)
- Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling (7 citations)
MDPs and planning · 4 papers
- Is a Good Representation Sufficient for Sample Efficient Reinforcement Learning? (30 citations)
- Observational Overfitting in Reinforcement Learning (16 citations)
- SVQN: Sequential Variational Soft Q-Learning Networks (7 citations)
- Learning the Arrow of Time for Problems in Reinforcement Learning (5 citations)
