US 11,741,344 B2
Custom convolutional neural network architectures for exposure detection
Eren Kursun, New York City, NY (US); and Hongda Shen, Metuchen, NJ (US)
Assigned to BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Dec. 9, 2019, as Appl. No. 16/707,916.
Prior Publication US 2021/0174167 A1, Jun. 10, 2021
Int. Cl. G06N 3/08 (2023.01); G06N 3/045 (2023.01); G10L 15/16 (2006.01)
CPC G06N 3/045 (2023.01) [G06N 3/08 (2013.01); G10L 15/16 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for generating custom neural network architectures, comprising:
a multi-layer device comprising one or more neural networks comprising one or more layers, wherein the system is configured to alter the one or more neural networks by:
customizing functional transformation of the one or more layers associated with a neural network architecture, wherein each of the one or more layers comprises a custom transformation function;
customizing interconnectivity of one or more layers associated with the neural network architecture, wherein the neural network architecture is associated with an application,
wherein the interconnectivity of one or more layers associated with the neural network architecture is customized via determining a custom connectivity function for a first layer of the one or more layers, and
wherein the custom connectivity function is based on the custom functional transformation of one or more preceding layers of the one or more layers that precede the first layer in the sequence;
dynamically altering the custom transformation function associated with each of the one or more layers, wherein the custom transformation function associated with each of the one or more layers is a function of time; and
generating a custom neural network architecture based on customizing the interconnectivity and the functional transformation of the one or more layers.