A quantum classification algorithm for classification incomplete patterns based on entanglement measure

Issue title: Mathematical Modelling in Computational and Life Sciences

Guest editors: Ahmed Farouk

Article type: Research Article

Authors: Abdel-Aty, Abdel-Haleem^{a; b; *} | Kadry, Heba^c | Zidan, Mohammed^d | Al-Sbou, Yazeed^e | Zanaty, E. A.^f | Abdel-Aty, Mahmoud^{c; d}

Affiliations: [a] Department of Physics, College of Sciences, University of Bisha, Bisha, Saudi Arabia | [b] Physics Department, Faculty of Science, Al-Azhar University, Assiut, Egypt | [c] Department of Mathematics and Computer Science, Faculty of Science, Sohag University, Sohag, Egypt | [d] University of Science and Technology, Zewail City of Science and Technology, Giza, Egypt | [e] Deanship of Research and Graduate Studies, Applied Science University, Manama, Kingdom of Bahrain | [f] Faculty of Computers and Information, Sohag University, Sohag, Egypt

Correspondence: [*] Corresponding author. Abdel-Haleem Abdel-Aty, E-mails: [email protected], [email protected].

Abstract: In this paper, a novel quantum classification algorithm that is based on competitive learning is presented to classify an input pattern that results from the failures of some sensors. As long as an incomplete pattern is presented to our model, the proposed algorithm performs the competitions between the neurons by applying some unitary transformations then measures the degree of entanglement using concurrence measure to find the winner class based on the winner-take-all technique. The proposed algorithm finds the most likely winning class label in between two binary competitive classes for an incomplete pattern presented to the proposed model. Because larger scale quantum computers are still in the lab, we studied the proposed algorithm on a case study.

Keywords: Quantum neural networks, incomplete patterns, quantum computing models

DOI: 10.3233/JIFS-179566

Journal: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 3, pp. 2809-2816, 2020