arXiv Science⌕ Search

arXiv subjects

D. Syvridis

Publications and source records attributed to D. Syvridis.

11 recordsLinked to original sources
arXiv2025

Learning kernels with quantum optical circuits

Support Vector Machines (SVMs) are a cornerstone of supervised learning, widely used for data classification. A central component of their success lies in kernel functions, which enable efficient computation of inner products in high-dimensional feature spaces. Recent years have seen growing interest in leveraging quantum circuits -- both qubit-based and quantum optical -- for computing kernel matrices, with ongoing research exploring potential quantum advantages. In this work, we investigate two classical techniques for enhancing SVM performance through kernel learning -- the Fisher criterion and quasi-conformal transformations -- and translate them into the framework of quantum optical circuits. Conversely, using the example of the displaced squeezed vacuum state, we demonstrate how established concepts from quantum optics can inspire novel perspectives and enhancements in SVM methodology. This cross-disciplinary approach highlights the potential of quantum optics to both inform and benefit from advances in machine learning.

quant-ph↗
arXiv2025

Applications of Hybrid Machine Learning Methods to Large Datasets: A Case Study

We combine classical and quantum Machine Learning (ML) techniques to effectively analyze long time-series data acquired during experiments. Specifically, we demonstrate that replacing a deep classical neural network with a thoughtfully designed Variational Quantum Circuit (VQC) in an ML pipeline for multiclass classification of time-series data yields the same classification performance, while significantly reducing the number of trainable parameters. To achieve this, we use a VQC based on a single qudit, and encode the classical data into the VQC via a trainable hybrid autoencoder which has been recently proposed as embedding technique. Our results highlight the importance of tailored data pre-processing for the circuit and show the potential of qudit-based VQCs.

quant-ph↗
arXiv2025

A Measurement Device Independent Quantum Key Distribution protocol in the service of three users

Quantum Key Distribution (QKD) is the only theoretically proven method for secure key distribution between two users. In this work, we propose and analyze a Measurement Device Independent (MDI) protocol designed to distribute keys among three users in a pairwise manner. Each user randomly selects a basis, encodes bit values in the phase of coherent states, and sends the resulting pulses to a central measurement unit (MU) composed of three beam splitters and three photon detectors. When the three pulses arrive simultaneously at the MU and under the condition of successful detection of photons, a key bit is distributed to at least one pair of users. This protocol extends the foundational phase-encoding MDI protocol introduced by [K. Tamaki, et al., Phys. Rev. A 85, 042307 (2012)] to three users, but this comes at the cost of introducing a systematic error in the implementation of the honest protocol.

quant-ph↗
arXiv2024

Enhancing the performance of Variational Quantum Classifiers with hybrid autoencoders

Variational Quantum Circuits (VQC) lie at the forefront of quantum machine learning research. Still, the use of quantum networks for real data processing remains challenging as the number of available qubits cannot accommodate a large dimensionality of data --if the usual angle encoding scenario is used. To achieve dimensionality reduction, Principal Component Analysis is routinely applied as a pre-processing method before the embedding of the classical features on qubits. In this work, we propose an alternative method which reduces the dimensionality of a given dataset by taking into account the specific quantum embedding that comes after. This method aspires to make quantum machine learning with VQCs more versatile and effective on datasets of high dimension. At a second step, we propose a quantum inspired classical autoencoder model which can be used to encode information in low latent spaces. The power of our proposed models is exhibited via numerical tests. We show that our targeted dimensionality reduction method considerably boosts VQC's performance and we also identify cases for which the second model outperforms classical linear autoencoders in terms of reconstruction loss.

quant-ph↗
arXiv2024

Real-time diagnostics on a QKD link via QBER Time Series Analysis

The integration of QKD systems in Metro optical networks raises challenges which cannot be completely resolved with the current technological status. In this work we devise a methodology for identifying different kind of impairments which may occur on the quantum channel during its transmission in an operational network. The methodology is built around a supervised ML pipeline which is using as input QBER and SKR time-series and requires no further interventions on the QKD system. The identification of impairments happens in real time and even though such information cannot reverse incidents, this can be valuable for users, operators and key management system.

quant-ph↗
arXiv2023

The Gell-Mann feature map of qutrits and its applications in classification tasks

Recent advancements in quantum hardware have enabled the realization of high-dimensional quantum states. This work investigates the potential of qutrits in quantum machine learning, leveraging their larger state space for enhanced supervised learning tasks. To that end, the Gell-Mann feature map is introduced which encodes information within an $8$-dimensional Hilbert space. The study focuses on classification problems, comparing Gell-Mann feature map with maps generated by established qubit and classical models. We test different circuit architectures and explore possibilities in optimization techniques. By shedding light on the capabilities and limitations of qutrit-based systems, this research aims to advance applications of low-depth quantum circuits.

quant-ph↗
arXiv2023

Classification of data with a qudit, a geometric approach

We propose a model for data classification using isolated quantum $d$-level systems or else qudits. The procedure consists of an encoding phase where classical data are mapped on the surface of the qudit's Bloch hyper-sphere via rotation encoding, followed by a rotation of the sphere and a projective measurement. The rotation is adjustable in order to control the operator to be measured, while additional weights are introduced in the encoding phase adjusting the mapping on the Bloch's hyper-surface. During the training phase, a cost function based on the average expectation value of the observable is minimized using gradient descent thereby adjusting the weights. Using examples and performing a numerical estimation of lossless memory dimension, we demonstrate that this geometrically inspired qudit model for classification is able to solve nonlinear classification problems using a small number of parameters only and without requiring entangling operations.

quant-ph↗
arXiv2020

Photonic Physical Unclonable Functions: From the Concept to Fully Functional Device Operating in the Field

The scope of this paper is to demonstrate a fully working and compact photonic Physical Unclonable Function (PUF) device capable of operating in real life scenarios as an authentication mechanism and random number generator. For this purpose, an extensive experimental investigation of a Polymer Optical Fiber (POF) and a diffuser as PUF tokens is performed and the most significant properties are evaluated using the proper mathematical tools. Two different software algorithms, the Random Binary Method (RBM) and Singular Value Decomposition (SVD), were tested for optimized key extraction and error correction codes have been incorporated for enhancing key reproducibility. By taking into consideration the limitations and overall performance derived by the experimental evaluation of the system, the designing details towards the implementation of a miniaturized, energy efficient and low-cost device are extensively discussed. The performance of the final device is thoroughly evaluated, demonstrating a long-term stability of 1 week, an operating temperature range of 50C, an exponentially large pool of unique Challenge-Response Pairs (CRPs), recovery after power failure and capability of generating NIST compliant true random numbers.

cs.CR↗