User Guide¶
This part of the documentation contains a user guide to a variety of Fairness, Accountability and Transparency algorithm. As opposed to the rest of the documentation, the user guide is focused on describing the theoretical and implementation (from a functional requirements point of view) aspects of FAT algorithms when applied to the three main components of a data processing pipeline: data, models and predictions.
Each entry in the user guide gives the description of the method, intended use of the method, available implementations in a variety of programming languages, best practices, advised usage and caveats, among many other listed properties.
Note
Additional learning resources are available on the FAT Forensics events website.
Fairness User Guide¶
The Fairness User Guide discusses techniques used for detecting, measuring and mitigating bias and unfairness in data and predictive algorithms.
Note
Coming Soon.
Accountability User Guide¶
The Accountability User Guide discusses safety, security, robustness and privacy aspects of data sets and predictive algorithms.
Note
Coming Soon.
Transparency User Guide¶
The Transparency User Guide tackles interpretability and explainability of data sets, predictive models and their predictions.
