Understanding the Resource Cost of Fully Homomorphic Encryption in Quantum Federated Learning

TL;DR

This paper evaluates the computational and communication overhead of applying Fully Homomorphic Encryption in Quantum Federated Learning, highlighting significant challenges and trade-offs between privacy and model performance.

Contribution

First implementation of a Quantum Convolutional Neural Network trained with CKKS-encrypted parameters in a federated setting, analyzing FHE overhead in QFL.

Findings

Memory and communication overhead are substantial with FHE.

Reducing model parameters decreases overhead but harms accuracy.

FHE deployment in QFL faces significant practical challenges.

Abstract

Quantum Federated Learning (QFL) enables distributed training of Quantum Machine Learning (QML) models by sharing model gradients instead of raw data. However, these gradients can still expose sensitive user information. To enhance privacy, homomorphic encryption of parameters has been proposed as a solution in QFL and related frameworks. In this work, we evaluate the overhead introduced by Fully Homomorphic Encryption (FHE) in QFL setups and assess its feasibility for real-world applications. We implemented various QML models including a Quantum Convolutional Neural Network (QCNN) trained in a federated environment with parameters encrypted using the CKKS scheme. This work marks the first QCNN trained in a federated setting with CKKS-encrypted parameters. Models of varying architectures were trained to predict brain tumors from MRI scans. The experiments reveal that memory and communication overhead remain substantial, making FHE challenging to deploy. Minimizing overhead requires reducing the number of model parameters, which, however, leads to a decline in classification performance, introducing a trade-off between privacy and model complexity.