公开(公告)号：US20220405529A1

公开(公告)日：2022-12-22

申请号：US17345730

申请日：2021-06-11

申请人： International Business Machines Corporation ,  REGENTS OF THE UNIVERSITY OF MICHIGAN

发明人： Mikhail Yurochkin ,  Debarghya Mukherjee ,  Moulinath Banerjee ,  Yuekai Sun ,  Sohini Upadhyay

IPC分类号： G06K9/62 ,  G06N3/08

摘要： The present invention provides techniques for learning Mahalanobis distance similarity metrics from data for individually fair machine learning models. In one aspect, a method for learning a fair Mahalanobis distance similarity metric includes: obtaining data with similarity annotations; selecting, based on the data obtained, a model for learning a Mahalanobis covariance matrix Σ; and learning the Mahalanobis covariance matrix Σ from the data using the model selected, wherein the Mahalanobis covariance matrix Σ fully defines the fair Mahalanobis distance similarity metric.

公开(公告)号：US20220318639A1

公开(公告)日：2022-10-06

申请号：US17213167

申请日：2021-03-25

申请人： International Business Machines Corporation ,  Regents of the University of Michigan

发明人： Sohini Upadhyay ,  Mikhail Yurochkin ,  Debarghya Mukherjee ,  Yuekai Sun ,  Amanda Ruth Garcia Bower ,  Seyed Hamid Eftekhari ,  Alexander Vargo ,  Fan Zhang

IPC分类号： G06N5/00 ,  G06K9/62

摘要： Obtain a first data set, a second data set, and a machine learning model. Construct a sensitive subspace of the first data set that defines a fair metric for distance among elements of the first data set. Fairly train the machine learning model on the first data set using a distributionally robust optimization approach based on the fair metric. Produce an individually fair set of labels by applying the fairly trained machine learning model to the second data set. Allocate a resource according to the individually fair set of labels.