公开(公告)号：US20230393960A1

公开(公告)日：2023-12-07

申请号：US17805377

申请日：2022-06-03

Applicant: Adobe Inc.

Inventor： Meghanath Macha Yadagiri ,  Anish Narang ,  Deepak Pai ,  Sriram Ravindran ,  Vijay Srivastava

IPC: G06F11/34 ,  G06K9/62 ,  G06N20/00

CPC classification number: G06F11/3452 ,  G06K9/6267 ,  G06K9/6263 ,  G06N20/00

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that control bias in machine learning models by utilizing a fairness deviation constraint to learn a decision matrix that modifies machine learning model predictions. In one or more embodiments, the disclosed systems generate, utilizing a machine learning model, predicted classification probabilities from a plurality of samples comprising a plurality of values for a data attribute. Moreover, the disclosed systems determine utilizing a decision matrix and the predicted classification probabilities, that the machine learning model fails to satisfy a fairness deviation constraint with respect to a value of the data attribute. In addition, the disclosed systems generate a modified decision matrix for the machine learning model to satisfy the fairness deviation constraint by selecting a modified decision threshold for the value of the data attribute.