Impacts of Code Optimization in Noise Reduction and Performance Improvement of Quantum Machine Learning Models for Anomaly Detection in Computer Networks

• Grant number: 23/12945-1
• Support Opportunities: Scholarships in Brazil - Master
• Start date: April 01, 2024
• End date: February 28, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Kelton Augusto Pontara da Costa
• Grantee: Inaê Soares de Figueiredo
• Host Institution: Faculdade de Ciências (FC). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil

Abstract

Cybersecurity is a growing concern, with privacy and protection against attacks being important contention points between organizations and customers. The application of modern technologies, such as machine learning models for anomaly detection, can be effective for privacy and security, and there is an urgency to understand how the application of modern technologies, like quantum computers, can bring even more advancements to these applications. Quantum computing is currently in a noisy era where the existing quantum computers cannot perform tasks without generating considerable amounts of error.This research project aims to analyze the performance of three popular machine learning models (Support Vector Machines, Neural Networks, and Decision Trees) in different quantum environments when applied to anomaly detection tasks using the UNSW-NB15 and CSE-CIC-IDS201 datasets. Furthermore, it will evaluate how applying classical and quantum code optimization techniques impacts the performance of the models. Each model will be implemented in simulators and real quantum computers, and the impact of the optimization on noise reduction will also be evaluated.