The obtained results show that the high accuracy of the solution

[tanjimajuha20]

Quantum convolutional neural networks ( QCNNs ) are a series of convolutional layers, or sequences of quantum operations, interleaved with pooling layers that together reduce the size of the stored information while preserving important features of the dataset.

Russian physicists from cayman islands phone number list the Quantum Information Technologies Laboratory at the MISiS University, together with colleagues from the Russian Quantum Center and Lomonosov Moscow State University, have presented for the first time a method for multi-class classification of images of 4 classes with high accuracy, based on the QCNN architecture. The researchers have improved the optimized structure of the quantum circuit and the quantum model of the perceptron — a mathematical or computer model of the brain's perception of information in the form of a logical circuit with transitions, associative and reactive elements, which is an elementary block of a neural network. The scientists tested the proposed classifier on various samples of four images of handwritten digits or photographs of clothes and shoes.

"We have implemented the proposed approach for the first time to solve the problem of classifying 4 classes of images - handwritten digits and clothing items, using eight qubits for data encoding and four auxiliary qubits. The corresponding machine learning procedure was implemented as a hybrid quantum-classical (variational) model. This approach can be implemented both on emulators and on real quantum processors. Quantum machine learning is one of the most interesting areas of application of quantum computers," explained Alexey Fedorov, head of the laboratory of quantum information technologies at NUST MISIS and RCC.
of the proposed method is similar to the accuracy of classical convolutional neural networks with a comparable number of trainable parameters.

In the future, scientists plan to make further optimization of the perceptron more efficient so that classification problems are solved significantly faster than by classical methods.

The study was carried out within the framework of the strategic direction "Quantum Internet" of the Priority 2030 Program, a grant from the Russian Science Foundation and the Roadmap for the Development of Quantum Computing.