Some fairness frameworks employ adversarial networks that actively discourage the model from encoding gender information. During training, an adversary attempts to predict gender from model outputs, and the main model learns to minimize this predictability, thus reducing gender bias in representations.

Some fairness frameworks employ adversarial networks that actively discourage the model from encoding gender information. During training, an adversary attempts to predict gender from model outputs, and the main model learns to minimize this predictability, thus reducing gender bias in representations.

Empowered by Artificial Intelligence and the women in tech community.
Like this article?

Interested in sharing your knowledge ?

Learn more about how to contribute.

Sponsor this category.