Module gradient 

Gradient storage and management for federated learning

This module provides:

• Gradient delta format (differences from base model)
• Gradient compression (sparsification, quantization, top-k)
• Gradient aggregation (averaging, weighted, momentum)
• Gradient verification (checksum, shape, outliers)

Structs

ClientInfo

Client information in federated learning

ConvergenceDetector

Convergence detection for federated learning

DifferentialPrivacy

Differential privacy for gradient protection

FederatedRound

Federated learning round

GradientAggregator

Gradient aggregation for federated learning

GradientCompressor

Gradient compression utilities

GradientDelta

Gradient delta (difference from base model)

GradientVerifier

Gradient verification utilities

ModelSyncProtocol

Model synchronization protocol for federated learning

PrivacyBudget

Privacy budget for differential privacy

QuantizedGradient

Quantized gradient (reduced precision)

SecureAggregation

Secure aggregation for federated learning (simplified)

SparseGradient

Sparse gradient representation

Enums

ClientState

Client state in federated learning

DPMechanism

Differential privacy mechanism types

GradientError

Errors that can occur during gradient operations

LayerGradient

Gradient for a single layer