by Tor H. Aasheim, Knut T. Hufthammer, Sølve Ånneland and Håavard Brynjulfsen and Marija Slavkovik
Abstract:
The use of artificial intelligence for decision making raises concerns about the societal impact of such systems. Traditionally, the product of a human decision-maker are governed by laws and human values. Decision-making is now being guided - or in some cases, replaced by machine learning classification which may reinforce and introduce bias. Algorithmic bias mitigation is explored as an approach to avoid this, however it does come at a cost: efficiency and accuracy. We conduct an empirical analysis of two off-the-shelf bias mitigation techniques from the AIF360 toolkit on a binary classification task. Our preliminary results indicate that bias mitigation is a feasible approach to ensuring group fairness.
Reference:
Bias mitigation with AIF360: A comparative study (Tor H. Aasheim, Knut T. Hufthammer, Sølve Ånneland and Håavard Brynjulfsen and Marija Slavkovik), In Norsk Informatikkonferanse, 2020, Awarded Best Paper.
Bibtex Entry:
@article{NIK2020-b,
abstract = {The use of artificial intelligence for decision making raises concerns
about the societal impact of such systems. Traditionally, the product
of a human decision-maker are governed by laws and human values.
Decision-making is now being guided - or in some cases, replaced
by machine learning classification which may reinforce and introduce
bias. Algorithmic bias mitigation is explored as an approach to avoid
this, however it does come at a cost: efficiency and accuracy. We
conduct an empirical analysis of two off-the-shelf bias mitigation
techniques from the AIF360 toolkit on a binary classification task. Our
preliminary results indicate that bias mitigation is a feasible approach
to ensuring group fairness.},
author = {Tor H. Aasheim and Knut T. Hufthammer and S{\o}lve {\AA}nneland and H{\aa}avard Brynjulfsen and Marija Slavkovik},
journal = {Norsk Informatikkonferanse},
note = {Awarded Best Paper},
title = {Bias mitigation with AIF360: A comparative study},
url = {https://ojs.bibsys.no/index.php/NIK/article/view/833},
year = {2020},
bdsk-url-1 = {https://ojs.bibsys.no/index.php/NIK/article/view/833}}