β AI & ML
Fair data, fair models
See it
A model only knows its examples
- If 90% of your photos are cats, the model gets great at cats and weak at dogs.
- Unbalanced data can make a model less accurate or unfair for the group that's under-represented.
- Always ask: who is missing from this data?
Data
who's in it?
Model
learns patterns
Outcome
fair for everyone?