Glossary term
Glossary term
Governance and Compliance
A situation in which sensitive attributes are present, but not included in the training data. Because sensitive attributes are often correlated with other attributes of one's data, a model trained with unawareness about a sensitive attribute could still have disparate impact with respect to that attribute, or violate other fairness constraints.
Created for this library
A bank's credit team applies unawareness to sensitive attributes in its model inputs and audits for proxy attributes that may reintroduce them.
A hiring-tech vendor uses unawareness to sensitive attributes as a baseline mitigation before considering more advanced fairness methods.
A health-tech startup applies unawareness to sensitive attributes in its triage model and follows up with proxy-attribute audits.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License