End-to-End Quantum Semantic Communication with Variational Quantum Neural Networks

Melek Krichen; Nikhitha Nunavath; Riccardo Bassoli; Soumaya Cherkaoui; Frank H. P. Fitzek

quant-phAug 30, 2026

End-to-End Quantum Semantic Communication with Variational Quantum Neural Networks

Authors: Melek Krichen, Nikhitha Nunavath, Riccardo Bassoli, Soumaya Cherkaoui, Frank H. P. Fitzek

Organizations: Polytechnique Montréal Montréal, Canada · TU Dresden Dresden, Germany

Abstract

Quantum-enabled learning is increasingly being explored for future communication and networking applications, including distributed sensing, Internet of Things (IoT), and distributed quantum computing. However, existing approaches often face challenges in scalable inference and efficient information processing. To address these limitations, this paper integrates quantum machine learning (QML) with quantum semantic communication (QSemCom) in an end-to-end learning framework. A classical dataset is first mapped to a low-dimensional feature representation and encoded by a variational quantum transmitter that learns task-relevant semantic features. The resulting features are transmitted through a quantum channel and processed by a receiver to perform a downstream classification task. The proposed framework is evaluated using the MNIST dataset and variational quantum neural networks (QNNs) under ideal and noisy quantum-channel conditions. First, a baseline model is trained over an ideal channel and evaluated under increasing depolarizing noise without retraining. A second set of experiments introduces a trainable receiver QNN and jointly optimizes the transmitter and receiver through end-to-end (E2E) training. The results demonstrate that jointly trained quantum semantic transceivers can adapt the transmitted representation to channel impairments and preserve task-relevant information, highlighting the potential of receiver-aware QML for robust inference in quantum semantic communication.

Figures & tables

Fig. 1 : Proposed E2E QSemCom framework. A classical semantic encoder first compresses the input into a low-dimensional latent representation, which is transformed into quantum states through angle encoding before being processed by the transmitter-side QNN. The encoded quantum state is transmitted through a quantum channel and recovered by a receiver-side VQNN. Finally, a classifier predicts the semantic class from the recovered semantic representation.

Layer	Configuration
Input	$28\times 28$ grayscale image
Flatten	$28\times 28\rightarrow 784$
Fully connected 1	$784\rightarrow 256$
Activation	ReLU
Fully connected 2	$256\rightarrow 128$
Activation	ReLU

TABLE I : Architecture of the classical semantic encoder.

Fig. 2 : Transmitter Variational Quantum Circuit (TX-QNN)

Component	Configuration
Quantum encoding	Angle encoding ( $R_{Y}$ rotations)
Latent space dimension	4
Number of qubits	4
Transmitter layers	3
Receiver layers	2
TX parameters	36

TABLE II : Quantum circuit architecture.

Baseline TX Training
Parameter	Value
Optimizer	AdamW
Batch size	64
Epochs	100
Learning rate	$10^{-4}$
Quantum channel	Perfect

TABLE III : Training configuration.

Layer	Configuration
Input	4 quantum features
Fully connected 1	$4\rightarrow 64$
Activation	ReLU
Dropout	0.35
Fully connected 2	$64\rightarrow 32$
Activation	ReLU

TABLE IV : Architecture of the semantic classifier.

Fig. 3 : Test loss, accuracy, and F1-score for TX-QNN under the perfect channel.

Fig. 4 : Best accuracy and F1-score for TX-QNN.

Fig. 5 : F1-score of TX-QNN under noisy channels.

Explore similar work

Tailor Made Embeddings for Quantum Machine Learning
Jun 24, 2026Aldo Lamarre, Dominik ŠafránekVariational Autoencoders Quantum Machine Learning
Department of Electrical Engineering, KAIST, Daejeon, South Korea · Center for Theoretical Physics of Complex Systems, Institute for Basic Science (IBS), Daejeon - 34126, Korea · Faculty of Mathematics and Physics, Charles University, Ke Karlovu 3, 121 16 Praha 2, Czech Republic
Quantum Global Variational Learning for Quantum Error Correction
Jun 7, 2026Shun Ryuzaki, Hideo MukaiQuantum Error Correction Variational Quantum Circuits
Computer Science Program, Graduate School of Science and Technology, Meiji University · Department of Computer Science, School of Science and Technology, Meiji University
Quantum Machine Learning for Industrial Applications
Jun 12, 2026Léo MonbroussouQuantum Convolutional Neural Networks Quantum Machine Learning
Sorbonne Université, Ecole Doctorale Informatique, Télécommunications et Electronique (n° 572), Computer Science

Jun 24, 2026, quant-phJ/K move · Enter open · S save

Tailor Made Embeddings for Quantum Machine Learning

Aldo Lamarre, Dominik Šafránek

Department of Electrical Engineering, KAIST, Daejeon, South Korea · Center for Theoretical Physics of Complex Systems, Institute for Basic Science (IBS), Daejeon - 34126, Korea · Faculty of Mathematics and Physics, Charles University, Ke Karlovu 3, 121 16 Praha 2, Czech Republic

Variational Autoencoders Quantum Machine Learning Quantum Neural Networks

Open paper AskPDF arXiv

Autoencoders transformed classical machine learning by solving the curse of dimensionality, enabling principled weight initialization and learning compact, structured representations. In this work, we extend this paradigm to quantum machine learning by introducing a variational autoencoder framework that learns task-specific quantum embeddings of classical data. We demonstrate that high-dimensional datasets, including ImageNet, can be compressed into a 13-qubit quantum representation while remaining reconstructable through a learned decoder. On MNIST (3 vs 5), our approach achieves 98.5% validation accuracy using a circuit-centric quantum classifier, within 1.2 percentage points of a classical neural network baseline (99.7%) and more than 30 percentage points above a naive amplitude-embedding approach. Unlike amplitude embeddings, which require full quantum state tomography for recovery, or angle embeddings, which generally rely on circuit inversion under restrictive assumptions, the proposed framework reconstructs the original data from only a polynomial number of measurements. The framework was further validated on IBM quantum hardware, confirming that the learned embeddings remain stable and reconstructable under real device noise.

Abstract

Explore similar work

Tailor Made Embeddings for Quantum Machine Learning

Quantum Global Variational Learning for Quantum Error Correction

Quantum Machine Learning for Industrial Applications