Intermediate · Research
Historical bias
Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.
TL;DR. Bias present in data reflecting past societal prejudices, inequalities, or stereotypes.
Technical Definition
Bias present in data reflecting past societal prejudices, inequalities, or stereotypes.
How it works
Historical bias refers to prejudices and inequalities that are embedded in the real world and subsequently captured in training data. For example, using loan default data from a past era might reflect discriminatory lending practices. This bias can cause ML models to perpetuate or even amplify these societal issues if not carefully addressed during development.
Related Concepts
- Fairness — The principle that AI systems should treat individuals and groups equitably.
- Bias (ethics/fairness) — Unfair prejudice or favoritism towards certain groups or things, which can influence data, system design, and user interactions.
- Responsible AI (RAI) — A holistic framework encompassing the ethical, legal, and societal implications of AI, promoting trustworthy and beneficial systems.