This repository contains the implementation of our Quantum Hybrid Diffusion Models paper, leveraging both quantum computing principles and classical diffusion models based on U-Net for advanced synthetic data generation tasks. The codebase is structured to facilitate experimentation with hybrid approaches, combining the strengths of quantum variational circuits and classical machine learning techniques.

• Quantum Variational Circuits: Integration of quantum circuits to enhance model expressiveness.
• Hybrid Architecture: Combines quantum and classical methodologies.
• Efficient Encoding: Techniques for embedding classical data into quantum states.
• Model Training and Evaluation: Scripts for training models and evaluating performance on synthetic and real-world datasets.

• configs/: Configuration files for various experimental setups.
• main.py: Main script to run the models.
• sampling.py: Functions for data sampling and preprocessing.
• train.py: Training routines for the hybrid models.
• unet.py: Implementation of the U-Net architecture.
• utils.py: Utility functions.

• Python 3.x

Clone the repository and install the required packages:

git clone https://github.com/NesyaLab/Quantum-Hybrid-Diffusion-Models.git
cd Quantum-Hybrid-Diffusion-Models
pip install -r requirements.txt

Run the main script with a specified configuration:

python main.py --config configs/config.yaml

If you use this code in your research, please cite the following reference:

@article{de2024towards,
  title={Towards Efficient Quantum Hybrid Diffusion Models},
  author={De Falco, Francesca and Ceschini, Andrea and Sebastianelli, Alessandro and Saux, Bertrand Le and Panella, Massimo},
  journal={arXiv preprint arXiv:2402.16147},
  year={2024}
}

We acknowledge the code from denoising-diffusion-flax by Yiyi Xu, which inspired our implementation of the U-Net and Denoising Diffusion parts. Special thanks to all contributors and the open-source community for their invaluable support.

We welcome contributions to enhance the functionality and performance of the models. Please submit pull requests or open issues for any improvements or bug fixes.

This project is licensed under the MIT License.

For detailed information on the theory and implementation of our models, please refer to our Quantum Hybrid Diffusion Models paper.