US 12,301,596 B2
Enhancing hybrid traditional neural networks with liquid neural network units for cyber security and offense protection
Elvis Nyamwange, Little Elm, TX (US)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on May 30, 2024, as Appl. No. 18/678,016.
Application 18/678,016 is a continuation of application No. 17/550,565, filed on Dec. 14, 2021, granted, now 12,028,355.
Prior Publication US 2024/0323203 A1, Sep. 26, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 9/40 (2022.01); G06N 3/045 (2023.01)
CPC H04L 63/1416 (2013.01) [G06N 3/045 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
by a computing device including one or more processors and memory:
determining, by a liquid neural network unit, that a consumer request to access data is a threat to an enterprise organization;
generating, by a readout and prediction unit, rules and algorithms to handle the threat;
encrypting, by a rules and training unit, the rules and the algorithms;
transmitting, from the rules and training unit and to a deep learning neural network, the encrypted rules and the encrypted algorithms;
training the deep learning neural network to handle the threat using the encrypted rules and the encrypted algorithms;
duplicating, by the deep learning neural network and in a black box computing environment, the threat; and
training the deep learning neural network to detect and handle threat techniques used in the duplicated threat.