Reinforcement learning is one form of AI that learns by interacting in the world. When interacting with humans, it can become especially effective.

This demo gives visitors the opportunity to see how an agent can learn to coordinate with humans without knowing about interacting with humans in advance, through the video game, Overcooked.

The demonstration is being run by Ziyan Wang (King's College London) and Xingzhou Lou (Chinese Academy of Sciences and King's College London). Visitors will be able to ask questions about reinforcement learning and how it is being used in the game Overcooked in this instance. Ziyan and Xingzhou are both members of the Cooperative AI Lab at the Department of Informatics, King's College London.

This event is in-person only. Registration is not required.

This event is part of the King's Festival of Artificial Intelligence, running from Wednesday 24 to Sunday 28 May 2023, which brings together speakers, exhibits, demos and screenings in an exciting programme of events. Take a look at the other events here.