NeurIPS 2019 accepted papers on Reinforcement learning
99 of the 1,429 papers accepted at NeurIPS 2019 (6.9%) are on Reinforcement learning, compared with 6.9% at NeurIPS 2018. They fall into 6 topics.
See NeurIPS 2019 in the interactive map
Submitting to NeurIPS? 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 · 46 papers
- Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction (351 citations)
- Learning to Perform Local Rewriting for Combinatorial Optimization (156 citations)
- When to Trust Your Model: Model-Based Policy Optimization (113 citations)
- DualDICE: Behavior-Agnostic Estimation of Discounted Stationary Distribution Corrections (111 citations)
- A Generalized Algorithm for Multi-Objective Reinforcement Learning and Policy Adaptation (106 citations)
- When to use parametric models in reinforcement learning? (101 citations)
- Neural Trust Region/Proximal Policy Optimization Attains Globally Optimal Policy (52 citations)
- Provably Global Convergence of Actor-Critic: A Case for Linear Quadratic Regulator with Ergodic Cost (48 citations)
- Reinforcement Learning with Convex Constraints (48 citations)
- A Geometric Perspective on Optimal Representations for Reinforcement Learning (46 citations)
- Policy Poisoning in Batch Reinforcement Learning and Control (40 citations)
- Fully Parameterized Quantile Function for Distributional Reinforcement Learning (38 citations)
and 34 more in the interactive map.
Visual and representation RL · 28 papers
- Language as an Abstraction for Hierarchical Deep Reinforcement Learning (97 citations)
- Curriculum-guided Hindsight Experience Replay (80 citations)
- Learning Reward Machines for Partially Observable Reinforcement Learning (67 citations)
- Learning Data Manipulation for Augmentation and Weighting (66 citations)
- Park: An Open Platform for Learning-Augmented Computer Systems (57 citations)
- Adaptive Auxiliary Task Weighting for Reinforcement Learning (56 citations)
- MCP: Learning Composable Hierarchical Control with Multiplicative Compositional Policies (54 citations)
- Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck (52 citations)
- Shaping Belief States with Generative Environment Models for RL (51 citations)
- Planning with Goal-Conditioned Policies (44 citations)
- Search on the Replay Buffer: Bridging Planning and Reinforcement Learning (39 citations)
- Hierarchical Reinforcement Learning with Advantage-Based Auxiliary Rewards (35 citations)
and 16 more in the interactive map.
MDPs and planning · 18 papers
- RUDDER: Return Decomposition for Delayed Rewards (105 citations)
- Towards Optimal Off-Policy Evaluation for Reinforcement Learning with Marginalized Importance Sampling (67 citations)
- Finite-time Analysis of Approximate Policy Iteration for the Linear Quadratic Regulator (38 citations)
- Provably Efficient Q-learning with Function Approximation via Distribution Shift Error Checking Oracle (35 citations)
- Non-Stationary Markov Decision Processes, a Worst-Case Approach using Model-Based Reinforcement Learning (22 citations)
- Near-Optimal Reinforcement Learning in Dynamic Treatment Regimes (19 citations)
- Explicit Explore-Exploit Algorithms in Continuous State Spaces (14 citations)
- Real-Time Reinforcement Learning (14 citations)
- A Regularized Approach to Sparse Optimal Policy in Reinforcement Learning (13 citations)
- Almost Horizon-Free Structure-Aware Best Policy Identification with a Generative Model (12 citations)
- Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies (12 citations)
- A Meta-MDP Approach to Exploration for Lifelong Reinforcement Learning (11 citations)
and 6 more in the interactive map.
Imitation learning · 5 papers
- Meta-Inverse Reinforcement Learning with Probabilistic Context Variables (34 citations)
- Goal-conditioned Imitation Learning (31 citations)
- Learner-aware Teaching: Inverse Reinforcement Learning with Preferences and Constraints (25 citations)
- SMILe: Scalable Meta Inverse Reinforcement Learning through Context-Conditional Policies (19 citations)
- Learning from Trajectories via Subgoal Discovery (9 citations)
