NeurIPS 2020 accepted papers on Reinforcement learning
164 of the 1,901 papers accepted at NeurIPS 2020 (8.6%) are on Reinforcement learning, compared with 6.9% at NeurIPS 2019. They fall into 6 topics.
See NeurIPS 2020 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 · 59 papers
- DISK: Learning local features with policy gradient (192 citations)
- Gradient Surgery for Multi-Task Learning (99 citations)
- Discovering Reinforcement Learning Algorithms (64 citations)
- Softmax Deep Double Deterministic Policy Gradients (47 citations)
- Reinforcement Learning with Combinatorial Actions: An Application to Vehicle Routing (46 citations)
- Variational Policy Gradient Method for Reinforcement Learning with General Utilities (44 citations)
- DisCor: Corrective Feedback in Reinforcement Learning via Distribution Correction (41 citations)
- A Self-Tuning Actor-Critic Algorithm (37 citations)
- Munchausen Reinforcement Learning (36 citations)
- Meta-Gradient Reinforcement Learning with an Objective Discovered Online (35 citations)
- A Finite-Time Analysis of Two Time-Scale Actor-Critic Methods (33 citations)
- Improving Sample Complexity Bounds for (Natural) Actor-Critic Algorithms (33 citations)
and 47 more in the interactive map.
Visual and representation RL · 52 papers
- Learning to Dispatch for Job Shop Scheduling via Deep Reinforcement Learning (236 citations)
- Reinforcement Learning with Augmented Data (231 citations)
- POMO: Policy Optimization with Multiple Optima for Reinforcement Learning (135 citations)
- Learning to Utilize Shaping Rewards: A New Approach of Reward Shaping (99 citations)
- Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model (64 citations)
- Multi-Task Reinforcement Learning with Soft Modularization (61 citations)
- Effective Diversity in Population Based Reinforcement Learning (49 citations)
- Improving Generalization in Reinforcement Learning with Mixture Regularization (47 citations)
- MDP Homomorphic Networks: Group Symmetries in Reinforcement Learning (45 citations)
- The NetHack Learning Environment (44 citations)
- AttendLight: Universal Attention-Based Reinforcement Learning Model for Traffic Signal Control (39 citations)
- Automatic Curriculum Learning through Value Disagreement (39 citations)
and 40 more in the interactive map.
MDPs and planning · 26 papers
- Reinforcement Learning with General Value Function Approximation: Provably Efficient Approach via Bounded Eluder Dimension (52 citations)
- On Reward-Free Reinforcement Learning with Linear Function Approximation (33 citations)
- PC-PG: Policy Cover Directed Exploration for Provable Policy Gradient Learning (31 citations)
- Breaking the Sample Size Barrier in Model-Based Reinforcement Learning with a Generative Model (27 citations)
- Model-based Reinforcement Learning for Semi-Markov Decision Processes with Neural ODEs (25 citations)
- On Efficiency in Hierarchical Reinforcement Learning (18 citations)
- Inverse Rational Control with Partially Observable Continuous Nonlinear Dynamics (16 citations)
- Task-agnostic Exploration in Reinforcement Learning (16 citations)
- Provably Efficient Reward-Agnostic Navigation with Linear Value Iteration (15 citations)
- Provably Efficient Reinforcement Learning with Kernel and Neural Function Approximations (14 citations)
- Adaptive Experimental Design with Temporal Interference: A Maximum Likelihood Approach (12 citations)
- Improved Sample Complexity for Incremental Autonomous Exploration in MDPs (9 citations)
and 14 more in the interactive map.
Imitation learning · 18 papers
- Reward-rational (implicit) choice: A unifying formalism for reward learning (69 citations)
- Language-Conditioned Imitation Learning for Robot Manipulation Tasks (64 citations)
- BAIL: Best-Action Imitation Learning for Batch Deep Reinforcement Learning (30 citations)
- An Imitation from Observation Approach to Transfer Learning with Dynamics Mismatch (20 citations)
- Toward the Fundamental Limits of Imitation Learning (20 citations)
- Error Bounds of Imitating Policies and Environments (17 citations)
- Strictly Batch Imitation Learning by Energy-based Distribution Matching (15 citations)
- f-GAIL: Learning f-Divergence for Generative Adversarial Imitation Learning (12 citations)
- Adversarial Soft Advantage Fitting: Imitation Learning without Policy Optimization (7 citations)
- Off-Policy Imitation Learning from Observations (6 citations)
- The MAGICAL Benchmark for Robust Imitation (6 citations)
- Policy Improvement via Imitation of Multiple Oracles (5 citations)
and 6 more in the interactive map.
Offline RL · 7 papers
- Conservative Q-Learning for Offline Reinforcement Learning (556 citations)
- MOPO: Model-based Offline Policy Optimization (238 citations)
- MOReL: Model-Based Offline Reinforcement Learning (152 citations)
- Critic Regularized Regression (93 citations)
- RL Unplugged: A Suite of Benchmarks for Offline Reinforcement Learning (9 citations)
- EcoLight: Intersection Control in Developing Regions Under Extreme Budget and Network Constraints (2 citations)
- Expert-Supervised Reinforcement Learning for Offline Policy Learning and Evaluation
RL from feedback · 2 papers
- Preference-based Reinforcement Learning with Finite-Time Guarantees (8 citations)
- When Counterpoint Meets Chinese Folk Melodies (5 citations)
