Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2605.16692 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911698211831808 |
|---|---|
| author | Evers, Thomas Meo, Cristian Bohmer, Wendelin Dauwels, Justin Oren, Yaniv |
| author_facet | Evers, Thomas Meo, Cristian Bohmer, Wendelin Dauwels, Justin Oren, Yaniv |
| contents | We introduce EfficientTDMPC, a sample-efficient model-based reinforcement learning method for continuous control built on the TD-MPC family of algorithms. Central to this family is a planner that aims to find an action sequence that maximizes the estimated return. The return is estimated using a learned model and value networks, each of which can introduce error. EfficientTDMPC proposes to reduce this error in two ways. First, it introduces an ensemble of dynamics models and averages the return estimates across those models and across different rollout depths. Second, it adds the option to apply an uncertainty penalty to the planner objective, yielding a planner that avoids actions with uncertain return estimates. It then adds practical improvements which increase buffer data freshness and reduce compute. Lastly, we find that our contributions enable EfficientTDMPC to benefit more from a higher update-to-data (UTD) ratio, further improving sample efficiency. To the best of our knowledge, in the low data regime of each benchmark, EfficientTDMPC achieves state-of-the-art (SOTA) in terms of sample efficiency on HumanoidBench-Hard and DMC hard, while matching SOTA on DMC easy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_16692 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EfficientTDMPC: Improved MPC Objectives for Sample-Efficient Continuous Control Evers, Thomas Meo, Cristian Bohmer, Wendelin Dauwels, Justin Oren, Yaniv Machine Learning Artificial Intelligence Robotics We introduce EfficientTDMPC, a sample-efficient model-based reinforcement learning method for continuous control built on the TD-MPC family of algorithms. Central to this family is a planner that aims to find an action sequence that maximizes the estimated return. The return is estimated using a learned model and value networks, each of which can introduce error. EfficientTDMPC proposes to reduce this error in two ways. First, it introduces an ensemble of dynamics models and averages the return estimates across those models and across different rollout depths. Second, it adds the option to apply an uncertainty penalty to the planner objective, yielding a planner that avoids actions with uncertain return estimates. It then adds practical improvements which increase buffer data freshness and reduce compute. Lastly, we find that our contributions enable EfficientTDMPC to benefit more from a higher update-to-data (UTD) ratio, further improving sample efficiency. To the best of our knowledge, in the low data regime of each benchmark, EfficientTDMPC achieves state-of-the-art (SOTA) in terms of sample efficiency on HumanoidBench-Hard and DMC hard, while matching SOTA on DMC easy. |
| title | EfficientTDMPC: Improved MPC Objectives for Sample-Efficient Continuous Control |
| topic | Machine Learning Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2605.16692 |