US 12,248,882 B2
Locating a decision boundary for complex classifier
James K. Baker, Maitland, FL (US)
Assigned to D5AI LLC, Maitland, FL (US)
Filed by D5AI LLC, Maitland, FL (US)
Filed on May 12, 2023, as Appl. No. 18/196,855.
Application 18/196,855 is a continuation of application No. 17/815,851, filed on Jul. 28, 2022, granted, now 11,687,788.
Application 17/815,851 is a continuation of application No. 17/810,778, filed on Jul. 5, 2022, granted, now 11,531,900, issued on Dec. 20, 2022.
Application 17/810,778 is a continuation of application No. 16/901,608, filed on Jun. 15, 2020, granted, now 11,410,050, issued on Aug. 9, 2022.
Application 16/901,608 is a continuation of application No. 16/645,710, abandoned, previously published as PCT/US2018/053519, filed on Sep. 28, 2018.
Claims priority of provisional application 62/564,754, filed on Sep. 28, 2017.
Prior Publication US 2023/0289611 A1, Sep. 14, 2023
Int. Cl. G06N 3/088 (2023.01); G06F 18/24 (2023.01); G06N 3/04 (2023.01); G06N 3/044 (2023.01); G06N 3/045 (2023.01); G06N 3/047 (2023.01); G06N 3/063 (2023.01); G06N 3/084 (2023.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01); G06F 12/0815 (2016.01); G06F 17/18 (2006.01); G06N 3/048 (2023.01)
CPC G06N 3/088 (2013.01) [G06F 18/24 (2023.01); G06N 3/04 (2013.01); G06N 3/044 (2023.01); G06N 3/045 (2023.01); G06N 3/047 (2023.01); G06N 3/063 (2013.01); G06N 3/084 (2013.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01); G06F 12/0815 (2013.01); G06F 17/18 (2013.01); G06N 3/048 (2023.01)] 36 Claims
OG exemplary drawing
 
1. A method for improving performance of a classifier, wherein the classifier comprises a deep neural network that is trained iteratively through machine learning, wherein the deep neural network of the classifier comprises an input layer, an output layer, and one or more inner layers, wherein the layers comprise interconnected nodes that process and transform input data through weighted connections, the method comprising:
computing, by the classifier implemented by a computer system, first and second scores for each of multiple data examples from a generator, wherein the first score is indicative of whether the data example belongs to a first data cluster and the second score is indicative of whether the data example belongs to a second data cluster, wherein the generator is trained with an objective such that, for each data example generated by the generator, the first and second scores computed by the classifier are equal;
back-propagating, by the computer system, for multiple data examples generated by the generator, partial derivatives from the classifier to obtain a vector, for each data example, that is orthogonal to a decision surface for the classifier;
detecting, by the computer system, a problem with the classifier based on changes in directions of the vectors; and
upon detecting a problem, adjusting, by the computer system, the classifier to reduce errors by the classifier caused by overfitting training data.