System to invoke update of machine learning models on edge computers
AssigneeHitachi, Ltd.
InventorsJeremy OSTERGAARD, Joydeep ACHARYA
Life of this patent
Estimating the term…
Abstract
Example implementations described herein involve systems and methods to substantially simultaneously orchestrate machine learning models over multiple resource constrained control edge devices, so that the overall system is more agile to changes in events and environmental conditions where the models have been deployed. The example implementations described herein involve multiple processes that when executed, determine a list of edge devices to be updated along with the corresponding models based on correlation.
Claims17
A method, comprising:
- for a model refresh event being triggered by an edge device from a plurality of edge devices:
- selecting ones of the plurality of edge devices to apply a model refresh based on a first correlation to the edge device;
- grouping the ones of the plurality of edge devices into a plurality of groups based on a second correlation between the ones of the plurality of edge devices;
- and
- selecting models to be applied to each of the plurality of groups.
A non-transitory computer readable medium, storing instructions for executing a process, the instructions comprising:
- for a model refresh event being triggered by an edge device from a plurality of edge devices:
- selecting ones of the plurality of edge devices to apply a model refresh based on a first correlation to the edge device;
- grouping the ones of the plurality of edge devices into a plurality of groups based on a second correlation between the ones of the plurality of edge devices;
- and
- selecting models to be applied to each of the plurality of groups.
An apparatus configured to manage a plurality of edge devices, the apparatus comprising:
- a processor, configured to:
- for a model refresh event being triggered by an edge device from a plurality of edge devices:
- select ones of the plurality of edge devices to apply a model refresh based on a first correlation to the edge device;
- group the ones of the plurality of edge devices into a plurality of groups based on a second correlation between the ones of the plurality of edge devices;
- and
- select models to be applied to each of the plurality of groups.
Dependencies are inferred from the claim text.