Home › Glossary › Safety › Fairness constraint

Advanced · Safety

Fairness constraint

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. An imposed restriction on an algorithm to ensure specific fairness criteria are met.

Technical Definition

An imposed restriction on an algorithm to ensure specific fairness criteria are met.

How it works

A fairness constraint is a rule or condition integrated into a machine learning model's training or output process to guarantee that certain fairness metrics are satisfied. This can involve modifying the loss function, applying post-processing to predictions, or directly incorporating mathematical constraints into the optimization problem.

Related Concepts

  • Bias (ethics/fairness) — Unfair prejudice or favoritism towards certain groups or things, which can influence data, system design, and user interactions.
  • Fairness metric — A quantifiable measure used to assess the fairness of a model's predictions across different groups.
  • Responsible AI (RAI) — A holistic framework encompassing the ethical, legal, and societal implications of AI, promoting trustworthy and beneficial systems.

Further Reading

  • Google ML Glossary