US 11,777,870 B1
Machine-learning (ML)-based systems and methods for maximizing resource utilization
Naga Vamsi Krishna Akkapeddi, Charlotte, NC (US); Maharaj Mukherjee, Poughkeepsie, NY (US); George Albero, Charlotte, NC (US); William August Stahlhut, The Colony, TX (US); Manu Kurian, Dallas, TX (US); and Kevin A. Delson, Woodland Hills, CA (US)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Jul. 8, 2022, as Appl. No. 17/860,149.
Int. Cl. G06N 20/00 (2019.01); G06F 16/27 (2019.01); H04L 29/06 (2006.01); H04L 47/783 (2022.01); H04L 47/762 (2022.01); H04L 47/70 (2022.01)
CPC H04L 47/783 (2013.01) [H04L 47/762 (2013.01); H04L 47/823 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A machine-learning (ML)-based digital communication system with maximized resource utilization, the system comprising:
a central server comprising a processor and a non-transitory memory storing computer executable instructions that, when run on the processor, are configured to cause the processor to transmit communications over a communication network;
a plurality of edge devices, wherein the plurality of edge devices are located at a logical edge of the communication network and communicate with each other and with the central server via the communication network; and
a machine-learning (ML) engine; wherein:
the central server is operated by an entity that is independent of the communication network;
the plurality of edge devices each comprise an authenticated software application that is provided by the entity; and
when a first one of the plurality of edge devices is directed to execute a task:
the ML engine calculates, for each of the plurality of edge devices, a predicted resource availability score, wherein the predicted resource availability score is reduced based on a future task scheduled to be performed on a device; and
the central server distributes the task among the plurality of edge devices based on the predicted resource availability score.