Active Learning

EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution

EmbodiedRSI：首個透過「假設引導」實現自主演化與持續學習的機器人框架

EmbodiedRSI introduces a Fast-Slow Dual-System that enables robots to autonomously propose hypotheses, conduct physical experiments, and co-evolve code and skills.