Reinforcement learning with prediction-based rewards
We’ve developed Random Network Distillation (RND), a prediction-based method for encouraging reinforcement learning agents to explore their environments through curiosity, which for the first time exceeds average human performance on Montezuma’s Revenge.
原文：https://www.a8q9.cn/en/article/reinforcement-learning-with-prediction-based-rewards-4b247e.html