ICML 2019 accepted papers on Reinforcement learning
77 of the 774 papers accepted at ICML 2019 (9.9%) are on Reinforcement learning, compared with 9.8% at ICML 2018. They fall into 5 topics.
See ICML 2019 in the interactive map
Submitting to ICML? 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 · 38 papers
- Off-Policy Deep Reinforcement Learning without Exploration (272 citations)
- Information-Theoretic Considerations in Batch Reinforcement Learning (96 citations)
- Action Robust Reinforcement Learning and Applications in Continuous Control (84 citations)
- Safe Policy Improvement with Baseline Bootstrapping (79 citations)
- Dynamic Weights in Multi-Objective Deep Reinforcement Learning (59 citations)
- Maximum Entropy-Regularized Multi-Goal Reinforcement Learning (48 citations)
- Understanding the Impact of Entropy on Policy Optimization (44 citations)
- Control Regularization for Reduced Variance Reinforcement Learning (39 citations)
- Taming MAML: Efficient unbiased meta-reinforcement learning (39 citations)
- CAB: Continuous Adaptive Blending for Policy Evaluation and Learning (32 citations)
- Hessian Aided Policy Gradient (31 citations)
- Diagnosing Bottlenecks in Deep Q-learning Algorithms (30 citations)
and 26 more in the interactive map.
Visual and representation RL · 21 papers
- Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables (214 citations)
- Quantifying Generalization in Reinforcement Learning (189 citations)
- A Deep Reinforcement Learning Perspective on Internet Congestion Control (162 citations)
- Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules (142 citations)
- SOLAR: Deep Structured Representations for Model-Based Reinforcement Learning (132 citations)
- Generative Adversarial User Model for Reinforcement Learning Based Recommendation System (96 citations)
- DeepMDP: Learning Continuous Latent Space Models for Representation Learning (59 citations)
- Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation (59 citations)
- Collaborative Evolutionary Reinforcement Learning (32 citations)
- EMI: Exploration with Mutual Information (29 citations)
- Neural Logic Reinforcement Learning (29 citations)
- Curiosity-Bottleneck: Exploration By Distilling Task-Specific Novelty (23 citations)
and 9 more in the interactive map.
MDPs and planning · 9 papers
- A Theory of Regularized Markov Decision Processes (116 citations)
- Provably Efficient Maximum Entropy Exploration (97 citations)
- Counterfactual Off-Policy Evaluation with Gumbel-Max Structural Causal Models (53 citations)
- Provably efficient RL with Rich Observations via Latent State Decoding (50 citations)
- Policy Certificates: Towards Accountable Reinforcement Learning (16 citations)
- The Value Function Polytope in Reinforcement Learning (12 citations)
- Kernel-Based Reinforcement Learning in Robust Markov Decision Processes (11 citations)
- Exploration Conscious Reinforcement Learning Revisited (4 citations)
- Reinforcement Learning in Configurable Continuous Environments (3 citations)
Imitation learning · 8 papers
- Imitating Latent Policies from Observation (55 citations)
- Provably Efficient Imitation Learning from Observation Alone (36 citations)
- Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations (30 citations)
- On the Feasibility of Learning, Rather than Assuming, Human Biases for Reward Inference (17 citations)
- Imitation Learning from Imperfect Demonstration (16 citations)
- Random Expert Distillation: Imitation Learning via Expert Policy Support Estimation (15 citations)
- Learning from a Learner (11 citations)
- Learning a Prior over Intent via Meta-Inverse Reinforcement Learning (8 citations)
