Study of quantum training algorithms for artificial neural networks

• Grant number: 13/11683-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: August 01, 2013
• End date: June 30, 2014
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Patrícia Rufino Oliveira
• Grantee: André Barbosa
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

This research project aims to study the state of the art, the fundamentals, principles and practices of Quantum Computing and to investigate how these elements can be incorporate to the architectures and training algorithms for Artificial Neural Networks (ANN). Research involving these issues has resulted in models known as Quantum Neural Networks (QNN), which is the main object of study of this proposal. In this work, it is intended to examine two distinct models of QNN, one of those proposed in 2003 by Dan Ventura and Bob Ricks and the other proposed in 2009 by Hong Xiao and Maojun Cao. In addition, these quantum approaches will be compared, via simulations, both among themselves and with respect to classical neural networks, with the aim of analyzing the advantages and disadvantages of the different models.