Timezone: »

Fair Infinitesimal Jackknife: Mitigating the Influence of Biased Training Data Points Without Refitting
Prasanna Sattigeri · Soumya Ghosh · Inkit Padhi · Pierre Dognin · Kush Varshney

Tue Nov 29 09:00 AM -- 11:00 AM (PST) @ Hall J #324

In consequential decision-making applications, mitigating unwanted biases in machine learning models that yield systematic disadvantage to members of groups delineated by sensitive attributes such as race and gender is one key intervention to strive for equity. Focusing on demographic parity and equality of opportunity, in this paper we propose an algorithm that improves the fairness of a pre-trained classifier by simply dropping carefully selected training data points. We select instances based on their influence on the fairness metric of interest, computed using an infinitesimal jackknife-based approach. The dropping of training points is done in principle, but in practice does not require the model to be refit. Crucially, we find that such an intervention does not substantially reduce the predictive performance of the model but drastically improves the fairness metric. Through careful experiments, we evaluate the effectiveness of the proposed approach on diverse tasks and find that it consistently improves upon existing alternatives.

Author Information

Prasanna Sattigeri (IBM Research)
Soumya Ghosh (MIT-IBM Watson AI Lab, IBM Research)
Inkit Padhi (IBM Research)
Pierre Dognin (IBM Research AI)
Kush Varshney (IBM Research)

More from the Same Authors