Theory of Fair Machine Learning

Project Information

FairML

Grant agreement ID: 101106768

DOI

10.3030/101106768

EC signature date 30 November 2023

Start date 1 August 2024

End date 31 July 2026

Funded under

• Marie Skłodowska-Curie Actions (MSCA)

Total cost

No data

EU contribution

€ 214 934,40

Coordinated by

KOBENHAVNS UNIVERSITET
Denmark

Project description

Balancing fairness and accuracy in developing ethical machine learning

Designing fair machine learning algorithms is not easy. The challenge is due to imbalanced training data and human biases. In this context, the MSCA-funded FairML project will develop oracle bounds of fairness restraints and a trade-off between fairness and accuracy. Using ensemble classifiers with majority vote, it aims to cancel out errors and biases. Additionally, it incorporates Pareto optimality, illegal bias tracing and long-term fairness capturing to comply with anti-subordination laws. By integrating learning theory tools, including causality and online learning, the project aims for moral responsibility in algorithm design. This enhances scientific understanding and aligns fairness concepts with legal counterparts, impacting applications in recruitment, criminal justice and lending, benefiting society at large.

Objective

Designing fair machine learning algorithms is challenging because the training data is often imbalanced and reflects (sometimes subconscious) biases of human annotators, leading to a possible propagation of biases into future decision-making. Besides, enforcing fairness usually leads to an inevitable deterioration of accuracy due to restrictions on the space of classifiers. In this project, I will address this challenge by developing oracle bounds of fairness restraints and a Pareto-dominated trade-off between fairness and accuracy using ensemble classifiers with the majority vote, to cancel out not only errors but also biases. I will also develop illegal bias tracing and long-term fairness capturing to comply with anti-subordination lawfully, using learning theory tools including causality and online learning for moral responsibility. The central objective of this proposal is to gain a theoretical understanding of fairness and to design machine learning algorithms that simultaneously improve both fairness and accuracy. The study is essential both for improved scientific understanding of fairness in machine learning models, and for the development of fairer algorithms for the numerous application domains, such as recruitment, criminal judging, or lending. Moreover, the project also takes interdisciplinary knowledge of economics and law into account to avoid fairness concepts in machine learning from being misaligned with their legal counterparts, enlarging the impact of machine learning applications and giving back to the wider community.

Fields of science (EuroSciVoc)

• social sciences › law

Keywords

• Fairness
• machine learning
• majority vote
• causality
• multi-armed bandits

Programme(s)

• HORIZON.1.2 - Marie Skłodowska-Curie Actions (MSCA) MAIN PROGRAMME
  

Topic(s)

• HORIZON-MSCA-2022-PF-01-01 - MSCA Postdoctoral Fellowships 2022
  

Funding Scheme

HORIZON-TMA-MSCA-PF-EF - HORIZON TMA MSCA Postdoctoral Fellowships - European Fellowships

Call for proposal

HORIZON-MSCA-2022-PF-01

Coordinator

KOBENHAVNS UNIVERSITET

• Net EU contribution: € 214 934,40
• Address: NORREGADE 10 ; 1165 KOBENHAVN ; Denmark
• Region: Danmark › Hovedstaden › Byen København
• Activity type: Higher or Secondary Education Establishments
• Total cost: No data