Personalized machine learning model management and deployment on edge devices

Patent metadata

Field	Value
Publication number	US-11178010-B2
Application number	US-202016800464-A
Country	US
Kind code	B2
Filing date	Feb 25, 2020
Priority date	Feb 25, 2020
Publication date	Nov 16, 2021
Grant date	Nov 16, 2021

Abstract

A computer-implemented method, a computer program product, and a computer system for personalized machine learning model management and deployment on edge devices. An edge server monitors activities performed on respective ones of edge devices. The edge server associates machine learning models in a model set with respective ones of the activities. The edge server predicts a next set of activities that are to be performed on the respective ones of the edge devices. The edge server deploys, on the respective ones of edge devices, machine learning models that are associated with the next set of the activities. Applications on the respective ones of the devices, which execute the next set of the activities, leverage the machine learning models that are associated with the next set of the activities.

First claim

What is claimed is:

1. A computer-implemented method for personalized machine learning model management and deployment on edge devices, the method comprising:
monitoring, by an edge server, activities performed on respective ones of edge devices;
associating, by the edge server, machine learning models in a model set with respective ones of the activities;
receiving, by the edge server, data associated with the activities, wherein the data associated with the activities is passed to the machine learning models in the model set by the respective ones of the edge devices, wherein the machine learning models in the model set are stored on a cloud server;
evaluating, by the edge server, the machine learning models in the model set, by evaluating at least one of: confidence of responses and feedback from users of the respective ones of the edge devices;
determining, by the edge server, which machine learning models are best associated with the respective ones of the activities;
predicting, by the edge server, a next set of activities that are to be performed on the respective ones of the edge devices;
deploying, by the edge server, on the respective ones of edge devices, machine learning models that are associated with the next set of the activities; and
wherein applications on the respective ones of the devices, executing the next set of the activities, leverage the machine learning models that are associated with the next set of the activities.

2. The computer-implemented method of claim 1 , further comprising:
receiving, by the edge server, from users of the respective ones of the edge devices, feedback on satisfaction with the application and the machine learning models that are associated with the next set of the activities, after the next set of the activities are concluded on the respective ones of the edge devices; and
updating, by the edge server, associations between the machine learning models in the model set with respective ones of the activities.

3. The computer-implemented method of claim 1 , further comprising:
discarding, by the edge devices, the machine learning models that are associated with the next set of the activities, after the next set of the activities are concluded on the respective ones of the edge devices.

4. The computer-implemented method of claim 1 , further comprising:
deploying, by the edge server, on the respective ones of edge devices, subsets of the machine learning models in the model set;
passing, by the respective ones of the edge devices, data associated with the activities to the machine learning models in the model set; and
evaluating, by the edge server, at least one of: confidence of responses and feedback from users of the respective ones of the edge devices.

5. The computer-implemented method of claim 1 , wherein the edge server determines the next set of the activities by capturing at least one of user actions, calendars, locations, social media interactions, and messaging.

6. A computer program product for personalized machine learning model management and deployment on edge devices, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by one or more processors, the program instructions executable to:
monitor, by an edge server, activities performed on respective ones of edge devices;
associate, by the edge server, machine learning models in a model set with respective ones of the activities;
receive, by the edge server, data associated with the activities, wherein the data associated with the activities is passed to the machine learning models in the model set by the respective ones of the edge devices, wherein the machine learning models in the model set are stored on a cloud server;
evaluate, by the edge server, the machine learning models in the model set, by evaluating at least one of: confidence of responses and feedback from users of the respective ones of the edge devices;
determine, by the edge server, which machine learning models are best associated with the respective ones of the activities;
predict, by the edge server, a next set of activities that are to be performed on the respective ones of the edge devices;
deploy, by the edge server, on the respective ones of edge devices, machine learning models that are associated with the next set of the activities; and
wherein applications on the respective ones of the devices, executing the next set of the activities, leverage the machine learning models that are associated with the next set of the activities.

7. The computer program product of claim 6 , further comprising the program instructions executable to:
receive, by the edge server, from users of the respective ones of the edge devices, feedback on satisfaction with the applications and the machine learning models that are associated with the next set of the activities, after the next set of the activities are concluded on the respective ones of the edge devices; and
update, by the edge server, associations between the machine learning models in the model set with respective ones of the activities.

8. The computer program product of claim 6 , further comprising the program instructions executable to:
discard, by the edge devices, the machine learning models that are associated with the next set of the activities, after the next set of the activities are concluded on the respective ones of the edge devices.

9. The computer program product of claim 6 , further comprising program instructions executable to:
deploy, by the edge server, on the respective ones of edge devices, subsets of the machine learning models in the model set;
pass, by the respective ones of the edge devices, data associated with the activities to the machine learning models in the model set; and
evaluate, by the edge server, at least one of: confidence of responses and feedback from users of the respective ones of the edge devices.

10. The computer program product of claim 6 , wherein the edge server determines the next set of the activities by capturing at least one of user actions, calendars, locations, social media interactions, and messaging.

11. A computer system for personalized machine learning model management and deployment on edge devices, the computer system comprising:
one or more processors, one or more computer readable tangible storage devices, and program instructions stored on at least one of the one or more computer readable tangible storage devices for execution by at least one of the one or more processors, the program instructions executable to:
monitor, by an edge server, activities performed on respective ones of edge devices;
associate, by the edge server, machine learning models in a model set with respective ones of the activities;
receive, by the edge server, data associated with the activities, wherein the data associated with the activities is passed to the machine learning models in the model set by the respective ones of the edge devices, wherein the machine learning models in the model set are stored on a cloud server;
evaluate, by the edge server, the machine learning models in the model set, by evaluating at least one of: confidence of responses and feedback from users of the respective ones of the edge devices;
determine, by the edge server, which machine learning models are best associated with the respective ones of the activities;
predict, by the edge server, a next set of activities that are to be performed on the respective ones of the edge devices;
deploy, by the edge server, on the respective ones of edge devices, machine learning models that are associated with the next set of the

Assignees

• IBM

Inventors

• Kozhaya Joseph
• Anderson Ryan
• Kwatra Shikhar
• Mazumder Sourav

Classifications

• H04L41/149

  for prediction of maintenance · CPC title

• H04L41/147

  for predicting network behaviour · CPC title

• H04L41/0895

  Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L41/0886 Primary

  Fully automatic configuration · CPC title