Convolutional Neural Networks (CNN) are one of the best current Machine Learning techniques to learn, detect and classify images. Researchers at CNRS, IRIF and Université Paris Diderot have described the first CNN implementation on quantum processors, with an advantage that could allow for bigger and faster CNN, thus many more applications.
The quantum CNN (QCNN) is a shallow circuit, reproducing completely the classical CNN, by allowing non linearities and pooling operations.