Little AI is a pedagogical game aimed at presenting the founding concepts of constructivist learning and developmental Artificial Intelligence. It primarily targets students in computer science and cognitive science but it can also interest the general public curious about these topics. It requires no particular scientific background; even children can find it entertaining. Professors can use it as a pedagogical resource in class or in online courses (e.g., MOOCs). The player presses buttons to control a simulated “baby robot” from the robot’s perspective; she cannot see the robot from the outside. The only information she receives from the environment is feedback from her actions; she cannot directly see the environment. She must learn, at the same time, the functioning of the robot’s body and the structure of the environment from regularities in the stream of commands and feedback. We argue that this situation is analogous to how infants engage in early-stage developmental learning (e.g., Piaget 1937).
- Little AI website: http://little-ai.com/
- The web version: https://github.com/pierreelliott/LittleAI-Web
- Survey about your experience playing Litte AI: https://clarolineconnect.univ-lyon1.fr/survey/survey/145/answer/form
To implement the new level n of Group g, create the following files (example with g=3 and nn=00):
- Level300.swift: the code of the level.
- Help3.plist: the instructions for the levels of Group 3.
- ImagineModel300.swift: The 3D simulation to show in the Replay Window.
Please contact us for more information.