US 11,949,567 B2
Safeguarding artificial intelligence-based network control
Lyndon Y. Ong, Sunnyvale, CA (US); David Côté, Gatineau (CA); Raghuraman Ranganathan, Bellaire, TX (US); and Thomas Triplet, Manotick (CA)
Assigned to Ciena Corporation, Hanover, MD (US)
Filed by Ciena Corporation, Hanover, MD (US)
Filed on Oct. 24, 2022, as Appl. No. 17/971,709.
Application 17/971,709 is a continuation of application No. 16/270,667, filed on Feb. 8, 2019, granted, now 11,483,212.
Prior Publication US 2023/0046886 A1, Feb. 16, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 41/16 (2022.01); G06F 30/27 (2020.01); G06N 7/00 (2023.01); G06N 20/00 (2019.01); G06N 20/20 (2019.01); H04L 12/46 (2006.01)
CPC H04L 41/16 (2013.01) [G06F 30/27 (2020.01); G06N 7/00 (2013.01); G06N 20/00 (2019.01); G06N 20/20 (2019.01); H04L 12/4641 (2013.01)] 20 Claims
OG exemplary drawing
 
9. A method comprising steps of:
obtaining data from a network having a plurality of network elements;
analyzing the data with one or more Machine Learning (ML) algorithms to determine one or more actions for network control where the one or more actions cause a configuration change across any of the plurality of network elements, wherein the network and the plurality of network elements operate at one or more layers including optical, time division multiplexing (TDM), packet, and combinations thereof;
analyzing the one or more actions that were determined by the ML algorithms to determine any risks associated therewith, prior to implementation of the one or more actions in the network; and
one of allowing, modifying, and blocking the one or more actions based on the risks to safeguard the network, wherein the modifying includes changing the one or more actions responsive to the one or more actions having unbounded uncertainties and there being overlap with a deterministic result.