According to the IT House on October 23, NVIDIA recently announced an AI system called Eureka, which is based on OpenAI's GPT-4 and can allow robots to perform more than 30 complex actions such as "turning the pen", "opening drawers", "holding scissors", "passing the ball with both hands" and so on. It is reported that NVIDIA Research Institute led the development of Eureka, an AI system that allows developers to cooperate with NVIDIA's own physics simulation software Isaac Gym for reinforcement learning (reference learning). Anima Anandkumar, senior director of NVIDIA's AI research department, believes that although "reinforcement learning" has made progress in the past 10 years, there are still many challenges, such as "reward design", which is still in the "trial and error" stage. At present, Nvidia announced Eureka, which is designed to perform difficult tasks, combined with generative AI and reinforcement learning algorithms, for the first attempt. Eureka is able to take over 80% of human experts, increasing the average robot training efficiency by more than 50%.