ICLR 2019 accepted papers on Reinforcement learning
63 of the 502 papers accepted at ICLR 2019 (12.5%) are on Reinforcement learning, compared with 10.7% at ICLR 2018. They fall into 4 topics.
See ICLR 2019 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 · 41 papers
- Large-Scale Study of Curiosity-Driven Learning (356 citations)
- Recurrent Experience Replay in Distributed Reinforcement Learning (252 citations)
- Exploration by random network distillation (245 citations)
- Episodic Curiosity through Reachability (184 citations)
- Learning to Adapt in Dynamic, Real-World Environments through Meta-Reinforcement Learning (101 citations)
- Diversity is All You Need: Learning Skills without a Reward Function (94 citations)
- Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic (84 citations)
- Learning Multi-Level Hierarchies with Hindsight (72 citations)
- From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Following (65 citations)
- Contingency-Aware Exploration in Reinforcement Learning (58 citations)
- Near-Optimal Representation Learning for Hierarchical Reinforcement Learning (58 citations)
- Policy Transfer with Strategy Optimization (56 citations)
and 29 more in the interactive map.
Policy gradient and actor-critic · 13 papers
- Reward Constrained Policy Optimization (254 citations)
- Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees (127 citations)
- CEM-RL: Combining evolutionary and gradient-based methods for policy search (93 citations)
- Variance Reduction for Reinforcement Learning in Input-Driven Environments (63 citations)
- Plan Online, Learn Offline: Efficient Learning and Exploration via Model-Based Control (61 citations)
- Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Search (39 citations)
- Soft Q-Learning with Mutual-Information Regularization (30 citations)
- A new dog learns old tricks: RL finds classic optimization algorithms (28 citations)
- Learning Self-Imitating Diverse Policies (23 citations)
- Recall Traces: Backtracking Models for Efficient Reinforcement Learning (23 citations)
- Two-Timescale Networks for Nonlinear Value Function Approximation (17 citations)
- Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications (6 citations)
and 1 more in the interactive map.
Imitation learning · 8 papers
- Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow (143 citations)
- Neural Probabilistic Motor Primitives for Humanoid Control (85 citations)
- Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning (65 citations)
- Directed-Info GAIL: Learning Hierarchical Policies from Unsegmented Demonstrations using Directed Information (37 citations)
- Sample Efficient Imitation Learning for Continuous Control (36 citations)
- Generative predecessor models for sample-efficient imitation learning (14 citations)
- Guiding Policies with Language via Meta-Learning (8 citations)
- Adversarial Imitation via Variational Inverse Reinforcement Learning (6 citations)
MDPs and planning · 1 paper
- Bayesian Policy Optimization for Model Uncertainty (4 citations)
