The special exhibition will take visitors through the history of AI, exploring key concepts and discovering its pitfalls and opportunities. By the end, visitors will have the chance to work with some AI tools themselves.

IMAGINARY’s exhibit demonstrates how AI uses Reinforcement Learning to learn through trial and error, much like humans and animals. In the interactive game, visitors navigate a robot through a maze. They can help the robot learn by placing rewards and adjusting its exploration ratio. The exhibit allows the creation of custom mazes, where the robot gradually learns to find the exit and avoid hazards while experimenting with bonus rewards to reinforce specific paths.