Fine-Tuned Variational Quantum Classifiers for Cyber Attacks Detection Based on Parameterized Quantum Circuits and Optimizers
Publication Date
1-1-2024
Document Type
Conference Proceeding
Publication Title
Proceedings - 2024 IEEE 48th Annual Computers, Software, and Applications Conference, COMPSAC 2024
DOI
10.1109/COMPSAC61105.2024.00144
First Page
1067
Last Page
1072
Abstract
Recent investigations into Quantum Machine Learning (QML) techniques have unveiled methodologies that accelerate training in established machine learning models to provide an alternative for capturing complex patterns. This study focuses on implementing a practical QML Algorithm, Variational Quantum Classification (VQC) for cybersecurity dataset so that detecting anomalies can be improved and faster by reducing number of attributes log2 M while training the model using Qiskit. Also, we study quantum algorithms to understand how it impacts on cyber datasets to detect anomalies in a improved way as it follows logarithms in the dimensionality reduction of quantum states which opens new horizons to quantum big data applications. Most importantly, we aim to also investigate the impact of various parameterized quantum circuits on VQC using quantum data as quantum states encoded by the cyber security dataset, NSL-KDD. In this research, we train VQC with various structures and parameters of quantum circuits as well as optimizers to adjust parameters of quantum circuits (ansatz) to minimize the objective function values so as to improve accuracy of the model in which quantum circuit, EfficientSU2, along with optimizer, COBYLA, outperforms the accuracy than other circuits and optimizers which shows great potential for improving cybersecurity systems. The research could effectively bridge in the gap between theory and implementation based quantum machine learning on cybersecurity systems.
Funding Number
1946442
Funding Sponsor
National Science Foundation
Keywords
optimizer, parameterized quantum circuit, quantum machine learning, variational quantum classifier
Department
Applied Data Science
Recommended Citation
Md Abdur Rahman, Mst Shapna Akter, Emily Miller, Bogdan Timofti, Hossain Shahriar, Mohammad Masum, and Fan Wu. "Fine-Tuned Variational Quantum Classifiers for Cyber Attacks Detection Based on Parameterized Quantum Circuits and Optimizers" Proceedings - 2024 IEEE 48th Annual Computers, Software, and Applications Conference, COMPSAC 2024 (2024): 1067-1072. https://doi.org/10.1109/COMPSAC61105.2024.00144
