A quantum convolutional neural network on NISQ devices

Abstract

Quantum machine learning is one of the most promising applications of quantum computing in the noisy intermediate-scale quantum (NISQ) era. We propose a quantum convolutional neural network(QCNN) inspired by convolutional neural networks (CNN), which greatly reduces the computing complexity compared with its classical counterparts, with O((log2M)6) basic gates and O(m2+e) variational parameters, where M is the input data size, m is the filter mask size, and e is the number of parameters in a Hamiltonian. Our model is robust to certain noise for image recognition tasks and the parameters are independent on the input sizes, making it friendly to near-term quantum devices. We demonstrate QCNN with two explicit examples. First, QCNN is applied to image processing, and numerical simulation of three types of spatial filtering, image smoothing, sharpening, and edge detection is performed. Secondly, we demonstrate QCNN in recognizing image, namely, the recognition of handwritten numbers. Compared with previous work, this machine learning model can provide implementable quantum circuits that accurately corresponds to a specific classical convolutional kernel. It provides an efficient avenue to transform CNN to QCNN directly and opens up the prospect of exploiting quantum power to process information in the era of big data.

Quantum cyber-physical systems

“…Another pending research topic is the associated optimization problems linked to multilayered networks, where cost functions become defined over the network. Finally, it is worth mentioning that the proposed approach is not a new design of quantum neural networks, such as the many proposals and ideas that can be found in the literature 58 , 59 . It can be seen as another perspective of quantum complex networks, which have been shown to be more vulnerable to some network attacks than others 60 , 61 .…”

Perfect NOT and conjugate transformations

“…A specific model of quantum programming of such kind is the measurement-based quantum computation [17]. Another example of such kind is the quantum convolutional neural network based on linear combination of unitaries [18]. As for entanglement purification, in addition to the pioneering work of Bennett et al [12], the recent experimental advancement [19] is worth noting.…”

Quantum-classical convolutional neural networks in radiological image classification

“…Given the success of CNNs in image classification tasks, different proposals [20]- [23] were developed to apply similar ideas on variational quantum algorithms. In [20] a quantum convolutional neural network (QCNN) with quantum convolutional, quantum pooling and quantum classification layers was proposed, with non-linearities achieved through quantum pooling layers.…”