Research map / Reinforcement learning / Policy gradient and actor-critic
Blending MPC & Value Function Approximation for Efficient 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 |
|---|---|---|---|
| Blend model-based MPC estimates with learned Q-values at every timestep | MPQ · CartpoleSwingup | average reward | Helped |
| Blend model-based MPC estimates with learned Q-values at every timestep | MPQ · SawyerPegInsertion | success rate | Helped |
| Blend model-based MPC estimates with learned Q-values at every timestep | MPQ · InHandManipulation | average reward | Helped |
| Use a high fixed λ of 1.0 or 0.95 (ablation) | MPQ · CartpoleSwingup | average reward | Hurt |
| Use λ below 0.6 (ablation) | MPQ · CartpoleSwingup | average reward | Hurt |
| Decay λ over training instead of keeping it fixed (ablation) | MPQ · CartpoleSwingup | convergence rate | Helped |
| Decay λ too fast or too slowly (ablation) | MPQ · CartpoleSwingup | average reward | Hurt |
| Use extremely biased dynamics (ablation) | MPQ · CartpoleSwingup | average reward | Hurt |
| Use bias factor b=1.0 for the pen dynamics (ablation) | MPQ · InHandManipulation | success rate | Hurt |
| Use λ blending across horizons rather than tune the discrete planning horizon (ablation) | MPQ · CartpoleSwingup | convergence rate | Helped |
| Use λ blending with very short planning horizons H=1 or 2 (ablation) | MPQ · CartpoleSwingup | average reward | Hurt |
Closest papers by meaning
- TD-MPC2: Scalable, Robust World Models for Continuous Control (ICLR 2024 · similarity 0.88)
- Bootstrapped Model Predictive Control (ICLR 2025 · similarity 0.88)
- Robust Model Based Reinforcement Learning Using $\mathcal{L}_1$ Adaptive Control (ICLR 2024 · similarity 0.87)
- Stochastic Q-learning for Large Discrete Action Spaces (ICML 2024 · similarity 0.87)
- Actor-Free Continuous Control via Structurally Maximizable Q-Functions (NeurIPS 2025 · similarity 0.87)
- Towards General-Purpose Model-Free Reinforcement Learning (ICLR 2025 · similarity 0.87)
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.