US 12,437,205 B2
Focused hyperparameter tuning using attribution
Wanlin Xie, Shelton, CT (US); and Mauro Joseph Sanchirico, III, Marlton, NJ (US)
Assigned to Lockheed Martin Corporation, Bethesda, MD (US)
Filed by Lockheed Martin Corporation, Bethesda, MD (US)
Filed on Apr. 13, 2023, as Appl. No. 18/299,942.
Prior Publication US 2024/0346331 A1, Oct. 17, 2024
Int. Cl. G06K 9/00 (2022.01); G06N 3/0985 (2023.01); G06V 10/776 (2022.01); G06V 10/82 (2022.01)
CPC G06N 3/0985 (2023.01) [G06V 10/776 (2022.01); G06V 10/82 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A system for training a neural network, the system comprising:
one or more memory units; and
a processor communicatively coupled to the one or more memory units, the processor configured to:
determine a plurality of test predictions by performing an inference process using a checkpointed learning model of the neural network and a plurality of test vectors, wherein the checkpointed learning model comprises a plurality of hyperparameters and a plurality of weights;
determine an attribution map by performing one or more attribution processes using the plurality of test predictions and the plurality of test vectors;
determine a score for each particular hyperparameter of the plurality of hyperparameters by analyzing the attribution map using an association classifier;
determine, based on the analysis by the association classifier, whether each particular hyperparameter of the plurality of hyperparameters should be frozen or tuned again; and
when it is determined that at least one particular hyperparameter should be tuned again, update the plurality of hyperparameters and the plurality of weights of the neural network.