Module federated 

Federated learning utilities.

This module provides:

• Differential privacy mechanisms (PrivacyBudget, DPMechanism, DifferentialPrivacy)
• Secure aggregation (SecureAggregation)
• Client lifecycle management (ClientState, ClientInfo)
• Round coordination (FederatedRound, ConvergenceDetector, ModelSyncProtocol)
• Standalone helpers: federated_average, clip_gradient_norm
• DistributedGradientAccumulator for peer-to-peer gradient exchange

Structs

ClientInfo

Client information in federated learning

ConvergenceConfig

Configuration for ConvergenceDetector.

ConvergenceDetector

Convergence detection for federated learning.

DifferentialPrivacy

Differential privacy for gradient protection

DistributedGradientAccumulator

Accumulates gradients from distributed peers via content-addressed storage.

FederatedRound

Federated learning round.

GossipModelSync

GossipSub-based model synchronisation protocol.

ModelSyncProtocol

Model synchronization protocol for federated learning

ModelUpdate

A model-update announcement broadcast via GossipSub.

PrivacyBudget

Privacy budget for differential privacy

RoundStats

Statistics snapshot for a completed federated round.

SecureAggregation

Secure aggregation for federated learning (simplified)

Enums

ClientState

Client state in federated learning

DPMechanism

Differential privacy mechanism types

FederatedError

Errors specific to the federated round lifecycle.