SLAC: Safe and Efficient Real-Robot RL via Unsupervised Simulation Pre-Training
SLAC is a reinforcement learning method that uses a low-fidelity simulator to pretrain a task-agnostic latent action space through unsupervised skill discovery, enabling efficient and safe real-world learning for high-degree-of-freedom robots. The approach achieves state-of-the-art performance on bimanual mobile manipulation tasks, learning contact-rich, whole-body behaviors in under an hour of real-world interaction without demonstrations or hand-crafted priors.
Why it matters: SLAC advances the feasibility of real-world reinforcement learning for complex robots by combining simulation-based pretraining with safe, sample-efficient real-world learning.
Full story at: arXiv Robotics ↗