|
Work item:
|
F.748.62 (ex F.DEC-CFML)
|
|
Subject/title:
|
Technical framework and requirements for device-edge-cloud collaborative federated machine learning
|
|
Status:
|
Consented on 2025-10-17 [Issued from previous study period]
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2025-10 (Medium priority)
|
|
Liaison:
|
ITU-T SG13
|
|
Supporting members:
|
China Mobile, China Telecom, BJTU China, ZTE China
|
|
Summary:
|
Federated machine learning (FML) allows collaborative model training via parameter exchange, ensuring privacy and compliance. A device–edge–cloud framework, with edge servers as intermediaries, enhances accuracy, resource balance, and scalability, requiring standardized protocols plus strategies for resource use, privacy, and aggregation. This Recommendation introduces the framework and requirements for device-edge-cloud collaborative FML, and specifies its concept, procedure, requirements and use cases.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2023-08-22 13:45:26
|
|
Last update:
2025-11-06 16:35:48
|
