AAAI 2023 accepted papers on Reinforcement learning
114 of the 2,023 papers accepted at AAAI 2023 (5.6%) are on Reinforcement learning, compared with 5.9% at AAAI 2022. They fall into 6 topics.
See AAAI 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Visual and representation RL · 51 papers
- SafeLight: A Reinforcement Learning Method toward Collision-Free Traffic Signal Control (49 citations)
- H-TSP: Hierarchically Solving the Large-Scale Traveling Salesman Problem (47 citations)
- Safe Reinforcement Learning via Shielding under Partial Observability (43 citations)
- Siamese-Discriminant Deep Reinforcement Learning for Solving Jigsaw Puzzles with Large Eroded Gaps (23 citations)
- Predicate Invention for Bilevel Planning (16 citations)
- Reward Design for an Online Reinforcement Learning Algorithm Supporting Oral Self-Care (15 citations)
- SplitNet: A Reinforcement Learning Based Sequence Splitting Method for the MinMax Multiple Travelling Salesman Problem (15 citations)
- AdapSafe: Adaptive and Safe-Certified Deep Reinforcement Learning-Based Frequency Control for Carbon-Neutral Power Systems (13 citations)
- Meta-Reinforcement Learning Based on Self-Supervised Task Representation Learning (13 citations)
- STL-Based Synthesis of Feedback Controllers Using Reinforcement Learning (13 citations)
- FanoutNet: A Neuralized PCB Fanout Automation Method Using Deep Reinforcement Learning (11 citations)
- Incremental Reinforcement Learning with Dual-Adaptive ε-Greedy Exploration (10 citations)
and 39 more in the interactive map.
Policy gradient and actor-critic · 31 papers
- Learning Control Policies for Stochastic Systems with Reach-Avoid Guarantees (26 citations)
- Evaluating Model-Free Reinforcement Learning toward Safety-Critical Tasks (23 citations)
- The Sufficiency of Off-Policyness and Soft Clipping: PPO Is Still Insufficient according to an Off-Policy Measure (20 citations)
- Augmented Proximal Policy Optimization for Safe Reinforcement Learning (19 citations)
- Off-Policy Proximal Policy Optimization (15 citations)
- PiCor: Multi-Task Deep Reinforcement Learning with Policy Correction (11 citations)
- Policy-Adaptive Estimator Selection for Off-Policy Evaluation (10 citations)
- Hybrid Learning with New Value Function for the Maximum Common Induced Subgraph Problem (9 citations)
- Bellman Meets Hawkes: Model-Based Reinforcement Learning via Temporal Point Processes (8 citations)
- CEM: Constrained Entropy Maximization for Task-Agnostic Safe Exploration (8 citations)
- Gradient-Adaptive Pareto Optimization for Constrained Reinforcement Learning (7 citations)
- Q-functionals for Value-Based Continuous Control (7 citations)
and 19 more in the interactive map.
MDPs and planning · 12 papers
- Safe Policy Improvement for POMDPs via Finite-State Controllers (10 citations)
- Provably Efficient Causal Model-Based Reinforcement Learning for Systematic Generalization (7 citations)
- Provably Efficient Primal-Dual Reinforcement Learning for CMDPs with Non-stationary Objectives and Constraints (6 citations)
- Robust Average-Reward Markov Decision Processes (6 citations)
- Formally Verified Solution Methods for Markov Decision Processes (5 citations)
- On the Challenges of Using Reinforcement Learning in Precision Drug Dosing: Delay and Prolongedness of Action Effects (4 citations)
- Planning and Learning for Non-markovian Negative Side Effects Using Finite State Controllers (4 citations)
- AlphaSnake: Policy Iteration on a Nondeterministic NP-Hard Markov Decision Process (Student Abstract) (1 citation)
- Computably Continuous Reinforcement-Learning Objectives Are PAC-Learnable (1 citation)
- Exploiting Multiple Abstractions in Episodic RL via Reward Shaping (1 citation)
- Model-Based Reinforcement Learning with Multinomial Logistic Function Approximation (1 citation)
- Tight Performance Guarantees of Imitator Policies with Continuous Actions
Offline RL · 11 papers
- Adaptive Policy Learning for Offline-to-Online Reinforcement Learning (24 citations)
- Towards Safe Mechanical Ventilation Treatment Using Deep Offline Reinforcement Learning (16 citations)
- Weighted Policy Constraints for Offline Reinforcement Learning (12 citations)
- Offline Quantum Reinforcement Learning in a Conservative Manner (7 citations)
- On Instance-Dependent Bounds for Offline Reinforcement Learning with Linear Function Approximation (5 citations)
- DARL: Distance-Aware Uncertainty Estimation for Offline Reinforcement Learning (2 citations)
- Flow to Control: Offline Reinforcement Learning with Lossless Primitive Discovery (2 citations)
- On the Sample Complexity of Vanilla Model-Based Offline Reinforcement Learning with Dependent Samples (2 citations)
- Behavior Estimation from Multi-Source Data for Offline Reinforcement Learning (1 citation)
- Model-Based Offline Reinforcement Learning with Local Misspecification
- Model-Based Offline Weighted Policy Optimization (Student Abstract)
Imitation learning · 8 papers
- The Perils of Trial-and-Error Reward Design: Misdesign through Overfitting and Invalid Task Specifications (48 citations)
- Reinforcement Learning for Branch-and-Bound Optimisation Using Retrospective Trajectories (20 citations)
- Offline Imitation Learning with Suboptimal Demonstrations via Relaxed Distribution Matching (9 citations)
- Unlabeled Imperfect Demonstrations in Adversarial Imitation Learning (9 citations)
- Misspecification in Inverse Reinforcement Learning (5 citations)
- Co-imitation: Learning Design and Behaviour by Imitation (4 citations)
- Expert Data Augmentation in Imitation Learning (Student Abstract) (1 citation)
- Learning Noise-Induced Reward Functions for Surpassing Demonstrations in Imitation Learning