Accelerating Visual Policy Learning with Sampling-Based Model Predictive Control
Sampling-Guided Policy Search (SGPS), which couples recurring action-target refinement by sampling-based model-predictive control with first-order policy optimization with first-order policy optimization, and shows that refinement improves policy learning beyond initialization and tracking alone.
Yi-Lang Liu, Hao-Xiang You, Qian Wang et al.
· 0 citations