Neural Net
Visitors interact with a physical representation of a neural network. Each node in the network represents a feature, and putting weight on that feature by pressing it or placing an object on it corresponds to placing a heavier weight on a particular feature in the network. In the example above, the neural network is a bird classifier. Placing more weight on a feature like “has wings” might cause birds like kiwis to be classified as “not a bird.” Visitors can explores different inputs and weights to see what outputs result.
Competencies: ML steps; representations; decision-making; learning from data; critically interpreting data
Design Considerations: social interaction; embodied interaction; opportunities to program; explainability