Mean Flow Policy Optimization (MFPO) uses few-step flow-based models for RL policies and achieves performance on par with or better than diffusion-based methods while substantially lowering training and inference time on MuJoCo and DeepMind Control Suite.
In: International Conference on Machine Learning, pp
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
Explainable DL integrated with DRL targets coherent structures sustaining turbulence, yielding 18.1% better net energy savings and superior generalization across Reynolds numbers and geometries than direct drag minimization.
citing papers explorer
-
Mean Flow Policy Optimization
Mean Flow Policy Optimization (MFPO) uses few-step flow-based models for RL policies and achieves performance on par with or better than diffusion-based methods while substantially lowering training and inference time on MuJoCo and DeepMind Control Suite.
-
Improving turbulence control through explainable deep learning
Explainable DL integrated with DRL targets coherent structures sustaining turbulence, yielding 18.1% better net energy savings and superior generalization across Reynolds numbers and geometries than direct drag minimization.