§ feed · storyline
Reinforcement learning with prediction-based rewards
OpenAI releases Random Network Distillation, a prediction-based reinforcement learning exploration method that surpasses average human performance on Montezuma's Revenge for the first time.
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.
§ sources1 publication · timeline below