The Golden Rule as a Heuristic to Measure the Fairness of Texts Using Machine Learning

TL;DR

This paper introduces a framework that applies the golden rule as a heuristic to evaluate the fairness of texts using machine learning, aiming to identify and mitigate unfair biases in language models.

Contribution

It operationalizes the golden rule in computational systems to measure and classify the fairness of sentences, addressing biases in word embeddings.

Findings

Proposes a method to categorize sentences as fair or unfair.

Suggests implementation strategies to reduce biases in word embeddings.

Provides a review of criticisms of the golden rule in computational context.

Abstract

In this paper we present a natural language programming framework to consider how the fairness of acts can be measured. For the purposes of the paper, a fair act is defined as one that one would be accepting of if it were done to oneself. The approach is based on an implementation of the golden rule (GR) in the digital domain. Despite the GRs prevalence as an axiom throughout history, no transfer of this moral philosophy into computational systems exists. In this paper we consider how to algorithmically operationalise this rule so that it may be used to measure sentences such as: the boy harmed the girl, and categorise them as fair or unfair. A review and reply to criticisms of the GR is made. A suggestion of how the technology may be implemented to avoid unfair biases in word embeddings is made - given that individuals would typically not wish to be on the receiving end of an unfair act, such as racism, irrespective of whether the corpus being used deems such discrimination as praiseworthy.