Research map / Reinforcement learning / Offline RL
Policy Regularization with Dataset Constraint for Offline Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Point-to-set dataset constraint (PRDC) added to TD3+BC-style policy update | TD3-based offline RL · D4RL Gym and AntMaze (18 tasks) | normalized score: SOTA on 13 of 18 tasks | Helped |
| Replace distribution/support constraint with nearest-neighbor dataset constraint | lineworld-hard and lineworld-superhard generalization tasks · lineworld generalization | task accomplished in all 5x10 evaluations: PRDC accomplishes all four; TD3+BC fails hard and superhard; BEAR fails medium onward | Helped |
| Average action of k nearest neighbors (k in {1,2,4}) instead of single nearest neighbor (ablation) | PRDC · hopper, halfcheetah, walker2d D4RL datasets | normalized score | Hurt |
| Small beta (0.1) for state-action distance (ablation) | PRDC · hopper-medium-v2 | normalized score | Hurt |
| Large beta (1000) for state-action distance (ablation) | PRDC · hopper-medium-v2 | normalized score | Hurt |
| Use KD-Tree for nearest neighbor search | PRDC · hopper-medium-v2 run time | run time | Helped |
Closest papers by meaning
- Iteratively Refined Behavior Regularization for Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.94)
- Adaptive Advantage-Guided Policy Regularization for Offline Reinforcement Learning (ICML 2024 · similarity 0.93)
- Regularizing a Model-based Policy Stationary Distribution to Stabilize Offline Reinforcement Learning (ICML 2022 · similarity 0.93)
- Learning to Trust Bellman Updates: Selective State-Adaptive Regularization for Offline RL (ICML 2025 · similarity 0.92)
- Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.91)
- Fewer May Be Better: Enhancing Offline Reinforcement Learning with Reduced Dataset (ICLR 2025 · similarity 0.91)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.