Nikolaos Schetakis

Mr. Nikolaos Schetakis

orcid

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 July 2022 in Scientific Reports
Verified
Journal Article
Open access
PDF file
Nikolaos SchetakisNikolaos Schetakis
Nikolaos Schetakis
DADavit Aghamalyan
Davit Aghamalyan
PGP. Griffin
P. Griffin
MBM. Boguslavsky
M. Boguslavsky
One of the most promising areas of research to obtain practical advantage is Quantum Machine Learning which was born as a result of cross-fertilisation of ideas between Quantum Computing and Classical Machine Learning. In this paper, we apply Quantum Machine Learning (QML) frameworks to improve binary classification models for noisy datasets which are prevalent in financial datasets. The metric we use for assessing the performance of our quantum classifiers is the area under the receiver operating characteristic curve AUC–ROC. By combining such approaches as hybrid-neural networks, parametric circuits, and data re-uploading we create QML inspired architectures and utilise them for the classification of non-convex 2 and 3-dimensional figures. An extensive benchmarking of our new FULL HYBRID classifiers against existing quantum and classical classifier models, reveals that our novel models exhibit better learning characteristics to asymmetrical Gaussian noise in the dataset compared to known quantum classifiers and performs equally well for existing classical classifiers, with a slight improvement over classical results in the region of the high noise.

10 December 2015 in Physical Review A
Verified
Journal Article
Open access
Changhyoup LeeChanghyoup Lee
Changhyoup Lee
CNChangsuk Noh
Changsuk Noh
Nikolaos SchetakisNikolaos Schetakis
Nikolaos Schetakis
DADimitris G. Angelakis
Dimitris G. Angelakis
We study the quantum transport of multiphoton Fock states in one-dimensional Bose-Hubbard lattices implemented in QED cavity arrays (QCAs). We propose an optical scheme to probe the underlying many-body states of the system by analyzing the properties of the transmitted light using scattering theory. To this end, we employ the Lippmann-Schwinger formalism within which an analytical form of the scattering matrix can be found. The latter is evaluated explicitly for the two-particle, two-site case which we use to study the resonance properties of two-photon scattering, as well as the scattering probabilities and the second-order intensity correlations of the transmitted light. The results indicate that the underlying structure of the many-body states of the model in question can be directly inferred from the physical properties of the transported photons in its QCA realization. We find that a fully resonant two-photon scattering scenario allows a faithful characterization of the underlying many-body states, unlike in the coherent driving scenario usually employed in quantum master-equation treatments. The effects of losses in the cavities, as well as the incoming photons' pulse shapes and initial correlations, are studied and analyzed. Our method is general and can be applied to probe the structure of any many-body bosonic model amenable to a QCA implementation, including the Jaynes-Cummings-Hubbard model, the extended Bose-Hubbard model, as well as a whole range of spin models.

3 June 2025 in Scientific Reports
Verified
Journal Article
Open access
Nikolaos SchetakisNikolaos Schetakis
Nikolaos Schetakis
PBPaolo Bonfini
Paolo Bonfini
NANegin Alisoltani
Negin Alisoltani
KBKonstantinos Blazakis
Konstantinos Blazakis
STSymeon I. Tsintzos
Symeon I. Tsintzos
EVEleni I. Vlahogianni
Eleni I. Vlahogianni
Accurate traffic forecasting plays a crucial role in modern Intelligent Transportation Systems (ITS), as it enables real-time traffic flow management, reduces congestion, and improves the overall efficiency of urban transportation networks. With the rise of Quantum Machine Learning (QML), it has emerged a new paradigm possessing the potential to enhance predictive capabilities beyond what classical machine learning models can achieve. In the present work we pursue a heuristic approach to explore the potential of QML, and focus on a specific transport issue. In particular, as a case study we investigate a traffic forecast task for a major urban area in Athens (Greece), for which we possess high-resolution data. In this endeavor we explore the application of Quantum Neural Networks (QNN), and, notably, we present the first application of quantum data re-uploading in the context of transport forecasting. This technique allows quantum models to better capture complex patterns, such as traffic dynamics, by repeatedly encoding classical data into a quantum state. Aside from providing a prediction model, we spend considerable effort in comparing the performance of our hybrid quantum-classical neural networks with classical deep learning approaches. We observe that, in fully connected network settings, hybrid quantum-classical models consistently underperform, with median scores approximately 10% worse than their purely classical counterparts across different configurations. In contrast, recursive architectures with data re-uploading show the opposite trend: hybrid models achieved up to 5% better median scores under comparable complexity settings. Additionally, these hybrid models converged in fewer training epochs, indicating improved training efficiency. Our results show that hybrid models achieve competitive accuracy with state-of-the-art classical methods, especially when the number of qubits and re-uploading blocks is increased. While the classical models demonstrate lower computational demands, we provide evidence that increasing the complexity of the quantum model improves predictive accuracy. These findings indicate that QML techniques, and specifically the data re-uploading approach, hold promise for advancing traffic forecasting models and could be instrumental in addressing challenges inherent in ITS environments.

4 April 2025 in IEEE Access
Verified
Journal Article
Open access
KBKonstantinos Blazakis
Konstantinos Blazakis
Nikolaos SchetakisNikolaos Schetakis
Nikolaos Schetakis
Mahmoud BadrMahmoud Badr
Mahmoud Badr
DADavit Aghamalyan
Davit Aghamalyan
KSKonstantinos Stavrakakis
Konstantinos Stavrakakis
GSGeorgios Stavrakakis
Georgios Stavrakakis
Electricity theft can lead to enormous economic losses and cause operational and security problems for electricity networks and utilities. Most current research has focused on electricity theft detection in the consumption sector. However, the high penetration rate of distributed generation (DG) can lead to an increase in power theft attacks in this sector via smart meter manipulation. This study is an extension of prior works focused on electricity theft detection in the consumption and generation domains of a smart grid environment with DG. This study proposes a novel electricity theft detection framework based on quantum machine learning (QML). The elegant field of QML has been used to demonstrate that data classification becomes more efficient in higher-dimensional spaces. An extensive numerical study was conducted to determine the type of QML architecture that can perform well and efficiently in electricity theft detection cases. The technique presented here has not yet been extensively studied in the domain of energy theft detection. Extensive experiments were conducted to assess this approach, and an accuracy of 0.87 was achieved with respect to the classical consumption domain, whereas an accuracy of 0.977 was achieved with respect to the net metering domain.

7 November 2013 in Journal of Physics B: Atomic, Molecular and Optical Physics
Verified
Journal Article
Closed access
Nikolaos SchetakisNikolaos Schetakis
Nikolaos Schetakis
TGThomas Grujic
Thomas Grujic
Stephen ClarkStephen Clark
Stephen Clark
DJDieter Jaksch
Dieter Jaksch
DADimitris Angelakis
Dimitris Angelakis

20 June 2023 in Remote Sensing
Verified
Journal Article
Open access
PDF file
MMMerope Manataki
Merope Manataki
Nikos PapadopoulosNikos Papadopoulos
Nikos Papadopoulos
Nikolaos SchetakisNikolaos Schetakis
Nikolaos Schetakis
AIAlessio Di Iorio
Alessio Di Iorio
This comparative study evaluates the performance of three popular deep learning architectures, AlexNet, VGG-16, and VGG-19, on a custom-made dataset of GPR C-scans collected from several archaeological sites. The introduced dataset has 15,000 training images and 3750 test images assigned to three classes: Anomaly, Noise, and Structure. The aim is to assess the performance of the selected architectures applied to the custom dataset and examine the potential gains of using deeper and more complex architectures. Further, this study aims to improve the training dataset using augmentation techniques. For the comparisons, learning curves, confusion matrices, precision, recall, and f1-score metrics are employed. The Grad-CAM technique is also used to gain insights into the models’ learning. The results suggest that using more convolutional layers improves overall performance. Further, augmentation techniques can also be used to increase the dataset volume without causing overfitting. In more detail, the best-obtained model was trained using VGG-19 architecture and the modified dataset, where the training samples were raised to 60,000 images through augmentation techniques. This model reached a classification accuracy of 94.12% on an evaluation set with 170 unseen data.