|
Work item:
|
TR.sgfdm (ex TR.sgfdcml)
|
|
Subject/title:
|
Fully Homomorphic Encryption (FHE) - based data aggregation in machine learning
|
|
Status:
|
Agreed on 2023-09-08 [Issued from previous study period]
|
|
Approval process:
|
Agreement
|
|
Type of work item:
|
Technical report
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
-
|
|
Liaison:
|
-
|
|
Supporting members:
|
-
|
|
Summary:
|
This Technical Report provides a guideline for secure data aggregation in machine learning (ML) while protecting input data. It focuses how Fully Homomorphic Encryption (FHE) works on data aggregations in machine learning. It first describes a general workflow on secure aggregation in ML and explains how FHE-based data aggregation in ML could satisfy a certain requirement. A general workflow is then given on FHE-based ML supporting data aggregation between more than two parties.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2020-03-27 16:31:09
|
|
Last update:
2023-10-17 12:12:35
|
