These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
FairTest
FairTest enables developers or auditing entities to discover and test for unwarranted associations between an algorithm’s outputs and certain user subpopulations identified by protected features.
FairTest works by learning a special decision tree, that splits a user population into smaller subgroups in which the association between protected features and algorithm outputs is maximized. FairTest supports and makes use of a variety of different fairness metrics each appropriate in a particular situation. After finding these so-called contexts of association, FairTest uses statistical methods to assess their validity and strength. Finally, FairTest retains all statistically significant associations, ranks them by their strength, and reports them as association bugs to the user.
About the tool
You can click on the links to see the associated tools
Developing organisation(s):
Tool type(s):
Objective(s):
Impacted stakeholders:
Country of origin:
Type of approach:
Maturity:
Target groups:
Target users:
Stakeholder group:
Geographical scope:
Required skills:
Use Cases
Would you like to submit a use case for this tool?
If you have used this tool, we would love to know more about your experience.
Add use case