Towards model-based bias mitigation in machine learning

TY - GEN

T1 - Towards model-based bias mitigation in machine learning

AU - Yohannis, Alfa

AU - Kolovos, Dimitris

N1 - Funding Information: This work has been funded through the York-Maastricht partnership’s Responsible Data Science by Design programme (https: //www.york.ac.uk/maastricht). We thank the Maastricht team for all their valuable contributions. Publisher Copyright: © 2022 ACM.

PY - 2022/10/23

Y1 - 2022/10/23

N2 - Models produced by machine learning are not guaranteed to be free from bias, particularly when trained and tested with data produced in discriminatory environments. The bias can be unethical, mainly when the data contains sensitive attributes, such as sex, race, age, etc. Some approaches have contributed to mitigating such biases by providing bias metrics and mitigation algorithms. The challenge is users have to implement their code in general/statistical programming languages, which can be demanding for users with little programming and fairness in machine learning experience. We present FairML, a model-based approach to facilitate bias measurement and mitigation with reduced software development effort. Our evaluation shows that FairML requires fewer lines of code to produce comparable measurement values to the ones produced by the baseline code.

AB - Models produced by machine learning are not guaranteed to be free from bias, particularly when trained and tested with data produced in discriminatory environments. The bias can be unethical, mainly when the data contains sensitive attributes, such as sex, race, age, etc. Some approaches have contributed to mitigating such biases by providing bias metrics and mitigation algorithms. The challenge is users have to implement their code in general/statistical programming languages, which can be demanding for users with little programming and fairness in machine learning experience. We present FairML, a model-based approach to facilitate bias measurement and mitigation with reduced software development effort. Our evaluation shows that FairML requires fewer lines of code to produce comparable measurement values to the ones produced by the baseline code.

KW - bias metrics

KW - bias mitigation

KW - generative programming

KW - machine learning

KW - model-driven engineering

UR - http://www.scopus.com/inward/record.url?scp=85141860329&partnerID=8YFLogxK

U2 - 10.1145/3550355.3552401

DO - 10.1145/3550355.3552401

M3 - Conference contribution

AN - SCOPUS:85141860329

T3 - Proceedings - 25th ACM/IEEE International Conference on Model Driven Engineering Languages and Systems, MODELS 2022

SP - 143

EP - 153

BT - Proceedings - 25th ACM/IEEE International Conference on Model Driven Engineering Languages and Systems, MODELS 2022

PB - Association for Computing Machinery, Inc

T2 - 25th ACM/IEEE International Conference on Model Driven Engineering Languages and Systems, MODELS 2022

Y2 - 23 October 2022 through 28 October 2022

ER -