What is it about?
Image classification is one of the most critical tasks in computer vision. Recently, numerous classification techniques based on quantum machine learning have been proposed, such as quanvolutional neural network (QNN) - a hybrid quantum-classical model which has the potential to process high-resolution images and outperform current image processing techniques. We investigate the use of entangled topologies in QNN to extract features more efficiently and propose a training strategy for the quantum part of the QNN model.
Featured Image
Photo by Fractal Hassan on Unsplash
Why is it important?
The results show that with a valid hyperparameter, our QNN achieves higher accuracy than CNN with a significantly smaller number of parameters.
Perspectives
This is my first publication about quantum machine learning, which is a hot research field today. It's also an achievement from my master's thesis. I hope this research gives useful experimental results in the quantum image processing field.
Hai Vu
Vietnam National University Ho Chi Minh City
Read the Original
This page is a summary of: Entangled topologies for quanvolutional neural networks in quantum image processing, December 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3628797.3628946.
You can read the full text:
Resources
Contributors
The following have contributed to this page
