Research map / Reinforcement learning / Policy gradient and actor-critic
Sample-Efficient Diffusion-based Reinforcement Learning with Critic Guidance
Shutong Ding, Zejia Zhong, Zhongyi Wang, Ke Hu, Bikang Pan, Jingya Wang, Ye Shi
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add critic guidance to target action generation (ablation) | CGPO · Ant-v3 | episodic return | Improved |
| use DSG-constrained guidance instead of naive guidance (ablation) | CGPO · Ant-v3 | episodic return | Improved |
| add DDQN target construction (ablation) | CGPO · Ant-v3 | episodic return | Improved |
| add truncated-quantile critic aggregation (ablation) | CGPO · Ant-v3 | episodic return | Improved |
| add value-calibrated base network for state weighting (ablation) | CGPO · Ant-v3 | episodic return | Improved |
| replace SAC with CGPO in HIL-SERL | HIL-SERL · Real-world peg-in-hole | success rate (%): 65 -> 80 | Improved |
| replace sampling-based candidate selection with critic-guided targets | diffusion policy RL (QVPO-style) · MuJoCo v3 (5 tasks) | best episodic return: Ant 6425.1 -> 7272.1 | Improved |
Closest papers by meaning
- Score Regularized Policy Optimization through Diffusion Behavior (ICLR 2024 · similarity 0.92)
- Diffusion Policy Policy Optimization (ICLR 2025 · similarity 0.92)
- Diffusion-based Reinforcement Learning via Q-weighted Variational Policy Optimization (NeurIPS 2024 · similarity 0.91)
- Prior-Guided Diffusion Planning for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.90)
- GenPO: Generative Diffusion Models Meet On-Policy Reinforcement Learning (NeurIPS 2025 · similarity 0.90)
- Learning Multimodal Behaviors from Scratch with Diffusion Policy Gradient (NeurIPS 2024 · similarity 0.89)
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.
