Learning Montezuma’s Revenge from a single demonstration
OpenAI demonstrates a major breakthrough in sample efficiency and reinforcement learning, showing agents can learn complex, long-horizon tasks from minimal human guidance. This is directly relevant to how AI systems will learn to perform real-world tasks at scale.
Why it ranks · · Score of 74,500 on Montezuma's Revenge from single demonstration · Jul 2 – 8, 2018
Read full story