US 11,769,056 B2
Synthetic data for neural network training using vectors
Sandipan Banerjee, Boston, MA (US); Rana el Kaliouby, Milton, MA (US); Ajjen Das Joshi, Arlington, MA (US); Survi Kyal, Chestnut Hill, MA (US); and Taniya Mishra, New York, NY (US)
Assigned to Affectiva, Inc., Boston, MA (US)
Filed by Affectiva, Inc., Boston, MA (US)
Filed on Dec. 29, 2020, as Appl. No. 17/136,083.
Claims priority of provisional application 63/083,136, filed on Sep. 25, 2020.
Claims priority of provisional application 63/071,401, filed on Aug. 28, 2020.
Claims priority of provisional application 62/955,493, filed on Dec. 31, 2019.
Claims priority of provisional application 62/954,819, filed on Dec. 30, 2019.
Claims priority of provisional application 62/954,833, filed on Dec. 30, 2019.
Prior Publication US 2021/0201003 A1, Jul. 1, 2021
Int. Cl. G06N 3/084 (2023.01); G06N 3/08 (2023.01); G06V 40/16 (2022.01); G06N 3/045 (2023.01); G06V 10/774 (2022.01); G06F 18/214 (2023.01); G06V 10/82 (2022.01)
CPC G06N 3/084 (2013.01) [G06F 18/214 (2023.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06V 40/16 (2022.01); G06V 40/165 (2022.01); G06V 40/169 (2022.01); G06V 40/174 (2022.01); G06V 40/179 (2022.01)] 25 Claims
OG exemplary drawing
 
1. A computer-implemented method for machine learning comprising:
obtaining facial images for a neural network training dataset;
encoding facial elements from the facial images into one or more vector representations of the facial elements using a machine learning neural network;
training a generative adversarial network (GAN) generator to provide one or more synthetic vectors based on the one or more vector representations, wherein a feature selector selects values from the one or more vector representations, the values being used as an input to the GAN, and wherein the one or more synthetic vectors enable avoidance of discriminator detection in the GAN;
generating additional synthetic vectors in the GAN, wherein the additional synthetic vectors avoid discriminator detection; and
further training the machine learning neural network, using the additional synthetic vectors that were generated by the GAN based on the selected values of the feature selector.