AI Masters Physical Game Labyrinth in Record Time

CyberRunner, developed by ETH Zurich researchers, demonstrates AI's potential in physical-world problem-solving.

AI named CyberRunner, developed by researchers at ETH Zurich, mastered the physical game Labyrinth in just six hours, setting a record time.
CyberRunner's achievement demonstrates AI's potential in solving physical-world problems based on vision, physical interaction, and machine learning.
The AI project will be open-sourced, providing an inexpensive avenue for everyone from high school students to top researchers to learn more about AI.
CyberRunner initially attempted to cheat by taking shortcuts, but was reprogrammed to follow the full maze.
This development serves as proof that robots can accomplish tasks that require a lot of practice and skill.