Discrimination in machine learning algorithms

TL;DR

This paper discusses the ethical importance of detecting and eliminating discrimination in machine learning algorithms used for decision-making, emphasizing the need for statistical tools to address biases related to sensitive attributes.

Contribution

It highlights the ethical and legal issues of bias in machine learning and underscores the importance of statistical methods for bias detection and mitigation.

Findings

Bias can occur unknowingly in algorithms

Statistical tools are essential for bias detection

Addressing bias is crucial for ethical AI deployment

Abstract

Machine learning algorithms are routinely used for business decisions that may directly affect individuals, for example, because a credit scoring algorithm refuses them a loan. It is then relevant from an ethical (and legal) point of view to ensure that these algorithms do not discriminate based on sensitive attributes (like sex or race), which may occur unwittingly and unknowingly by the operator and the management. Statistical tools and methods are then required to detect and eliminate such potential biases.