ETD-HUB

Inherent Bias in Image Dataset Labelling

Asked: 9 months, 2 weeks ago By: Deleuze Views: 209 Technology: Facial Recognition Bias

To what extent do a persons own biases 'contaminate' the fairness of a dataset during the labelling process? Are such ethical issues an inherent problem - i.e. is it ever possible to separate a persons own perspectives and biases from this process?

2 Answers

Answered: 9 months, 2 weeks ago By: Agavornik
Kate Crawford and Trevor Paglen, “Excavating AI: The Politics of Training Sets for Machine Learning (September 19, 2019) https://excavating.ai
Answered: 9 months, 2 weeks ago By: Hollytibble
Yeah this is a really important consideration when you don't have a 'gold standard' label, like 'death' but something more objective.

Your Answer

Login to add your answer!

We’d love to hear your thoughts — share a meaningful answer by logging in.