One-network Adversarial Fairness
AAAI19, Thirty-Third AAAI Conference on Artificial Intelligence (2019) Honolulu, Hawaii, USA
There is currently a great expansion of the impact of machine learning algorithms on our lives, prompting the need for ob- jectives other than pure performance, including fairness. Fair- ness here means that the outcome of an automated decision- making system should not discriminate between subgroups characterized by sensitive attributes such as gender or race. Given any existing differentiable classifier, we make only slight adjustments to the architecture including adding a new hidden layer, in order to enable the concurrent adversarial op- timization for fairness and accuracy. Our framework provides one way to quantify the tradeoff between fairness and accu- racy, while also leading to strong empirical performance.