direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Benchmark Website with Measurements for Fairness Frameworks

Fair and Discrimination-aware machine learning has recently received a lot of attention both from legal institutions as well as from the computer science community. A lot of research has been done on discrimination-aware classification and other machine learning tasks. However most of the approaches are not really comparable because there are datasets missing as well as benchmarks. A benchmarking website has been set up and already contains a few fairness measures that can be used to compare different approaches. 

This bachelor thesis aims to further develop the measurements framework and setup a proper library which can be easily downloaded and used by researchers to compare their methods against each other. 

Fairness and Discrimination measures to be implemented can be found here: https://arxiv.org/pdf/1511.00148.pdf

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions

Contact

Meike Zehlike
+49 (30) 314-27998
Room E-N 173