This repository contains the implementation of a Quantum Support Vector Machine (QSVM) enhanced by a genetic algorithm to optimize the feature map selection process. The project aims to explore and improve the performance of quantum machine learning by leveraging the power of quantum kernels, which are difficult to evaluate classically, and optimizing them using genetic algorithms.
- QSVM Implementation: The core of the project is a Quantum Support Vector Machine that uses quantum feature maps to classify data into two or more classes.
- Genetic Algorithm for Feature Map Optimization: A novel approach is introduced to optimize the selection of quantum feature maps using genetic algorithms. This method emulates natural selection to iteratively improve the accuracy of the QSVM on complex datasets.
- Implementation in Qiskit: The project leverages IBM's Qiskit framework for building and simulating quantum circuits.
- Dataset Used: The Iris dataset is primarily used to demonstrate the QSVM's performance, with additional artificially generated datasets to test the algorithm's robustness on non-linear data.