Managing Internet of Things network traffic using federated machine learning

A system, comprising:

• one or more computers;
• and
• one or more computer-readable media storing instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
• receiving, from each of at least a subset of a set of edge devices, machine learning parameters of a local machine learning model trained by the edge device, the machine learning parameters received from an edge device being results of machine learning training of the local machine learning model trained (i) by the edge device based on local network traffic processed by the edge device and (ii) to classify the network traffic as Internet of Things (IoT) traffic or non-IoT traffic;
• generating, using the machine learning parameters received from the subset of edge devices, a plurality of composite machine learning models to classify network traffic processed by edge devices as IoT traffic or non-IoT traffic, wherein generating, using the machine learning parameters received from the subset of edge devices, the plurality of composite machine learning models to classify the network traffic processed by edge devices as IoT traffic or non-IoT traffic comprises:
• training respective composite machine learning models for different groups of users or edge devices;
• and
• providing, for each of the different groups of users or edge devices, the corresponding composite machine learning model to the set of edge devices in the group or edge devices of users in the group, for use in classifying network traffic processed by the edge devices as IoT traffic or non-IoT traffic.

One or more non-transitory computer-readable media storing instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:

• receiving, from each of at least a subset of a set of edge devices, machine learning parameters of a local machine learning model trained by the edge device, the machine learning parameters received from an edge device being results of machine learning training of the local machine learning model trained (i) by the edge device based on local network traffic processed by the edge device and (ii) to classify the network traffic as Internet of Things (IoT) traffic or non-IoT traffic;
• generating, using the machine learning parameters received from the subset of edge devices, a plurality of composite machine learning models to classify network traffic processed by edge devices as IoT traffic or non-IoT traffic, wherein generating, using the machine learning parameters received from the subset of edge devices, the plurality of composite machine learning models to classify the network traffic processed by edge devices as IoT traffic or non-IoT traffic comprises:
• training respective composite machine learning models for different groups of users or edge devices;
• and
• providing, for each of the different groups of users or edge devices, the corresponding composite machine learning model to edge devices in the group or edge devices of users in the group, for use in classifying network traffic processed by the edge devices as IoT traffic or non-IoT traffic.