Best in Automated Facial Recognition

$100 CAD cash / person Data used to train automated facial recognition models are often not equally inclusive of all demographics. Thus, these models are more error-prone for some users (often those not as included in the engineering design/coding process) than others. How can we de-bias automated facial recognition models for greater accuracy for all users?

Winner will also receive an MLH winner pin and the opportunity to have their MLH Fellowship application expedited should they choose to apply.