Table of Contents - M.3387 (03/2024) - Management requirements for federated machine learning systems

1	Scope
2	References
3	Definitions
	3.1	Terms defined elsewhere
	3.2	Terms defined in this Recommendation
4	Abbreviations and acronyms
5	Conventions
6	Overview
7	Scenario of federated machine learning system management
8	Requirements for basic management domain
	8.1	Requirements for system property configuration
	8.2	Requirements for node permission configuration
	8.3	Requirements for node property configuration
	8.4	Requirements for service operation management
	8.5	Requirements for service request management
9	Requirements for model management domain
	9.1	Requirements for initial model deployment
	9.2	Requirements for learning algorithm management
	9.3	Requirements for aggregation mechanism management
	9.4	Requirements for model training management
	9.5	Requirements for model quality evaluation
10	Requirements for data management domain
	10.1	Requirements for data flow management
	10.2	Requirements for data secure storage
	10.3	Requirements for data format normalization
	10.4	Requirements for related data retrieval
	10.5	Requirements for data encrypted transmission
Appendix I – A use case example of using FMLMS to manage FMLM training for road anomaly detection services in the Internet of vehicles
	I.1	Introduction
	I.2	Cloud-edge-terminal collaborative FML architecture
	I.3	The management process of FMLS in road anomaly detection model training
Bibliography