US 12,147,906 B2
Localization-based test generation for individual fairness testing of artificial intelligence models
Diptikalyan Saha, Bangalore (IN); Aniya Aggarwal, New Delhi (IN); and Sandeep Hans, New Delhi (IN)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Apr. 26, 2021, as Appl. No. 17/239,857.
Prior Publication US 2022/0343179 A1, Oct. 27, 2022
Int. Cl. G06N 5/01 (2023.01); G06N 20/00 (2019.01)
CPC G06N 5/01 (2023.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
obtaining at least one artificial intelligence model and training data related to the at least one artificial intelligence model;
identifying one or more boundary regions associated with the at least one artificial intelligence model based at least in part on results of processing at least a portion of the training data using the at least one artificial intelligence model, wherein identifying one or more boundary regions comprises generating, using at least one extraction algorithm, at least one artificial intelligence-based decision tree surrogate model of the at least one artificial intelligence model and identifying one or more regions pertaining to at least one boundary of the at least one artificial intelligence-based decision tree surrogate model;
generating, in accordance with at least one of the one or more identified boundary regions, one or more synthetic data points for inclusion with the training data;
executing one or more fairness tests on the at least one artificial intelligence model using at least a portion of the one or more generated synthetic data points and at least a portion of the training data; and
modifying, based at least in part on results of executing the one or more fairness tests, one or more parameters of the at least one artificial intelligence model;
wherein the method is carried out by at least one computing device.