vFedProtoQNAS: Prototype-Guided Personalized Quantum Neural Architecture Search for Virtual Federated Learning

Seok Bin Son; Samuel Yen-Chi Chen; Soohyun Park; Joongheon Kim

cs.AIOct 1, 2026

vFedProtoQNAS: Prototype-Guided Personalized Quantum Neural Architecture Search for Virtual Federated Learning

Authors: Seok Bin Son, Samuel Yen-Chi Chen, Soohyun Park, Joongheon Kim

Organizations: Korea University Seoul, Korea · Wells Fargo New York, USA · Sookmyung Women’s University Seoul, Korea

Abstract

Quantum federated learning (QFL) has emerged as a promising approach for collaboratively training compact quantum neural networks (QNNs) over distributed private data on resource-constrained devices. However, differences in device capabilities make a single shared QNN architecture unsuitable for all clients. While personalized quantum neural architecture search (QNAS) allows each client to select a device-specific QNN, averaging parameters across structurally different QNN architectures mixes semantically inconsistent circuit operations. To address this, prototype-guided personalized QNAS for virtual FL (vFedProtoQNAS) is proposed, where model parameters are never aggregated across clients and federated collaboration is achieved through class-wise prototype sharing. Each client independently searches and trains a client-specific QNN, computes class-wise local prototypes from latent representations, and refines them using global prototypes from the server as federated semantic anchors. Experiments demonstrate that vFedProtoQNAS improves accuracy by 3.70% over FedAvg and enhances class-consistent representation alignment.

Figures & tables

Figure 1: The concept of virtual federated learning.

Figure 2: The overall framework of vFedProtoQNAS.

Figure 3: Prototype margin improvement before and after prototype-guided refinement. A positive margin indicates that a latent representation is closer to the correct class prototype than to any incorrect prototype.

$\beta_{\mathrm{proto}}$	Accuracy	Precision	Recall	F1-score
$0.05$	$\mathbf{91.25{\pm}0.53}$	$\mathbf{91.30{\pm}0.52}$	$\mathbf{91.25{\pm}0.53}$	$91.23{\pm}0.54$
$0.10$	$\mathbf{91.25{\pm}0.74}$	$91.29{\pm}0.71$	$\mathbf{91.25{\pm}0.74}$	$\mathbf{91.24{\pm}0.74}$
$0.20$	$91.19{\pm}0.66$	$91.23{\pm}0.66$	$91.19{\pm}0.66$	$91.18{\pm}0.68$
$0.50$	$90.94{\pm}0.38$	$90.98{\pm}0.35$	$90.94{\pm}0.38$	$90.93{\pm}0.37$

Table 1: Sensitivity analysis of vFedProtoQNAS with respect to the prototype alignment weight

\beta_{\mathrm{proto}}

using MNIST(0-3). Results are reported as mean

\pm

std over five seeds.

$\alpha$	Accuracy	Precision	Recall	F1-score
$0.5$	$84.32{\pm}11.77$	$82.67{\pm}16.72$	$84.32{\pm}11.77$	$82.64{\pm}15.48$
$1.0$	$91.15{\pm}0.76$	$\mathbf{91.26{\pm}0.67}$	$91.15{\pm}0.76$	$91.15{\pm}0.75$
$10.0$	$\mathbf{91.16{\pm}0.93}$	$91.24{\pm}0.89$	$\mathbf{91.16{\pm}0.93}$	$\mathbf{91.17{\pm}0.92}$

Table 2: Sensitivity analysis of vFedProtoQNAS under different non-IID levels using MNIST(0-3). Results are reported as mean

\pm

std over five seeds. A smaller Dirichlet concentration parameter

\alpha

indicates stronger data heterogeneity.

Method	Accuracy	Precision	Recall	F1-score
FedAvg	$87.55{\pm}2.38$	$87.96{\pm}2.26$	$87.55{\pm}2.38$	$87.46{\pm}2.39$
FixedQNN LocalOnly	$90.31{\pm}1.01$	$90.54{\pm}0.88$	$90.31{\pm}1.01$	$90.25{\pm}1.04$
vFedProtoQNAS w/o QNAS	$90.01{\pm}0.76$	$90.21{\pm}0.71$	$90.01{\pm}0.76$	$89.92{\pm}0.83$
RandomQNAS + FedProto	$90.17{\pm}0.92$	$90.32{\pm}0.87$	$90.17{\pm}0.92$	$90.15{\pm}0.90$
PersonalQNAS	$90.45{\pm}0.70$	$90.59{\pm}0.66$	$90.45{\pm}0.70$	$90.44{\pm}0.71$
vFedProtoQNAS (Ours)	$\mathbf{91.25{\pm}0.74}$	$\mathbf{91.29{\pm}0.71}$	$\mathbf{91.25{\pm}0.74}$	$\mathbf{91.24{\pm}0.74}$

Table 3: Comparison with baselines under the default setting

\beta_{\mathrm{proto}}=0.10

. Results are reported as mean

\pm

std over five seeds.

Explore similar work

Q-ANCHOR: Federated Quantum Learning with ZNE-guided Correction
May 28, 2026Hoang M. Ngo, Quan Nguyen, Wanli Xing +1Federated Learning Federated Averaging
Department of Computer & Information Science & Engineering University of Florida · Frost Institute for Data Science and Computing University of Miami
A Drift Stable Quantum Federated Learning for Intelligent Services
Jul 22, 2026Shanika Iroshi Nanayakkara, Shiva Raj PokhrelFederated Learning Quantum Neural Networks
School of IT, Deakin University, VIC 3125, Burwood, Australia
QFedAgent: Quantum-Enhanced Personalized Federated Learning for Multi-Agent Activity Recognition
Jul 2, 2026Quoc Bao Phan, Tuy Tan NguyenFederated Learning Hybrid Quantum-Classical Pipeline
Department of Electrical and Computer Engineering FAMU-FSU College of Engineering, Florida State University Tallahassee, FL 32310, USA

May 28, 2026, cs.LGJ/K move · Enter open · S save

Q-ANCHOR: Federated Quantum Learning with ZNE-guided Correction

Hoang M. Ngo, Quan Nguyen, Wanli Xing, My T. Thai

Department of Computer & Information Science & Engineering University of Florida · Frost Institute for Data Science and Computing University of Miami

Federated Learning Federated Averaging Correction

Open paper AskPDF arXiv

Quantum Federated Learning (QFL) offers a promising framework to train quantum models across distributed clients while keeping data strictly local. Due to its simplicity and low communication overhead, Federated Averaging (FedAvg) is the standard aggregation choice in QFL literature. However, deploying QFL on practical hardware exposes a severe double-drift phenomenon: the global model is simultaneously derailed by client drift from non-IID data and hardware bias from noisy quantum gradient estimates. In this work, we first analyze the convergence of FedAvg under these realistic conditions, mathematically demonstrating that quantum hardware bias creates a persistent error floor that standard averaging cannot correct. To overcome this limitation, we propose Q-ANCHOR, a quantum-aware federated aggregation architecture that anchors server updates with zero-noise extrapolation while applying stateful client correction to suppress both client drift and hardware-induced bias. Our convergence theory proves that Q-ANCHOR successfully mitigates classical client drift while actively reducing the hardware-bias floor. Experimental results demonstrate that Q-ANCHOR achieves significantly more stable training than conventional FL baselines.

Abstract

Explore similar work

Q-ANCHOR: Federated Quantum Learning with ZNE-guided Correction

A Drift Stable Quantum Federated Learning for Intelligent Services

QFedAgent: Quantum-Enhanced Personalized Federated Learning for Multi-Agent Activity Recognition