Glossary term
Glossary term
Governance and Compliance
Showing partiality to one's own group or own characteristics. If testers or raters consist of the machine learning developer's friends, family, or colleagues, then in-group bias may invalidate product testing or the dataset.
In-group bias is a form of group attribution bias. See also out-group homogeneity bias.
See Fairness: Types of bias in Machine Learning Crash Course for more information.
Created for this library
A people-analytics team flags in-group bias as a risk when reviewing performance ratings shared by managers who are also close colleagues.
A research lab includes in-group bias in its annotator training so labelers are reminded not to be lenient with familiar groups.
A risk modeling team's review checklist includes in-group bias as a sanity check when model errors cluster within a known relationship.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License