US 11,843,515 B2
Peer risk benchmarking using generative adversarial networks
Qihong Shao, Clyde Hill, WA (US); David John Zacks, Vancouver (CA); and Xinjun Zhang, San Jose, CA (US)
Assigned to CISCO TECHNOLOGY, INC., San Jose, CA (US)
Filed by Cisco Technology, Inc., San Jose, CA (US)
Filed on Dec. 16, 2022, as Appl. No. 18/067,068.
Application 18/067,068 is a continuation of application No. 17/704,449, filed on Mar. 25, 2022, granted, now 11,533,241.
Application 17/704,449 is a continuation of application No. 17/077,073, filed on Oct. 22, 2020, granted, now 11,316,750, issued on Apr. 26, 2022.
Prior Publication US 2023/0118857 A1, Apr. 20, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 15/173 (2006.01); H04L 41/14 (2022.01); H04L 41/12 (2022.01); H04L 41/147 (2022.01); H04L 41/5067 (2022.01); H04L 43/045 (2022.01); H04L 43/06 (2022.01); H04L 43/0817 (2022.01); H04L 43/55 (2022.01)
CPC H04L 41/145 (2013.01) [H04L 41/12 (2013.01); H04L 41/147 (2013.01); H04L 41/5067 (2013.01); H04L 43/045 (2013.01); H04L 43/06 (2013.01); H04L 43/0817 (2013.01); H04L 43/55 (2022.05)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
obtaining customer data for a first network, wherein the customer data indicates a role of one or more network devices in the first network and a plurality of risk reports corresponding to the one or more network devices, wherein each risk report is associated with one of a plurality of dimensions of risk that include a recommended configuration of hardware or software of the first network;
generating a network profile image by processing the plurality of risk reports;
generating a synthetic network profile image from the network profile image, wherein the synthetic network profile image does not include the customer data;
evaluating a second network using the synthetic network profile image using a second network profile image generated by processing a second plurality of risk reports corresponding to a second plurality of dimensions of risk of the second network by comparing the synthetic network profile image to the second network profile image; and
identifying differences between the first network and the second network based on a comparison of the plurality of dimensions of risk to the second plurality of dimensions of risk.