Distributed and Secure Kernel-Based Quantum Machine Learning

This repository contains the implementation of the algorithms presented in the paper titled "Distributed and Secure Kernel-Based Quantum Machine Learning". It provides all the necessary code to reproduce the experiments described in the paper, including dataset loading, centralized and distributed computations, and experiment results.

Repository Structure

• library.py: This file contains all the core functions required to implement both the centralized and the distributed and secure kernel-based quantum machine learning algorithm as described in the paper. It serves as the backbone of the implementation, handling tasks such as kernel computation, noise management, and distributed processing.

• Datasets Folders: Each folder in the repository corresponds to a specific dataset used in the experiments. The folders include:

  • Data Loading Scripts: Scripts to load and preprocess the respective datasets.
  • Distributed QML Scripts: Code to run the distributed kernel-based quantum machine learning algorithm on the dataset.
  • Experimental Results: Results obtained from running the algorithm on the dataset, including kernel matrices and performance metrics.

• other_kernels: This folder contains the code for implementing the polynomial, RBF, and Laplacian kernels as proposed in the paper.

Citation

Please consider citing our work if it is beneficial to your research.

@article{
swaminathan2025distributed,
title={Distributed and Secure Kernel-Based Quantum Machine Learning},
author={Arjhun Swaminathan and Mete Akg{\"u}n},
journal={Transactions on Machine Learning Research},
issn={2835-8856},
year={2025},
url={https://openreview.net/forum?id=3jdI0aEW3k},
note={}
}

License

This project is released under the MIT License. See LICENSE for details.

Contact for Questions

arjhun.swaminathan@uni-tuebingen.de, mete.akguen@uni-tuebingen.de