Projeto evolutivo de circuitos combinacionais quânticos

• Autor(a) principal: Miranda Filho, Fernando Tenório de
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações do Mackenzie
• Texto Completo: http://dspace.mackenzie.br/handle/10899/26585
• topic: computação quântica; circuitos quânticos; matriz unitária; algoritmos evolutivos; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::MATEMATICA DA COMPUTACAO::MODELOS ANALITICOS E DE SIMULACAO; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO::LINGUAGENS DE PROGRAMACAO; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO::ENGENHARIA DE SOFTWARE

While advances in quantum hardware occur in modest steps, simulators running on classical computers allow a slightly faster advance in the construction of quantum algorithms. Given a unitary matrix that performs a given operation, obtaining the equivalent quantum circuit is a non-trivial task that can be modeled as a search problem. A generic approach that makes possible the automatic synthesis of these circuits would allow not only the study and development of new algorithms, but also to evaluate the cost-bene t related to the size and complexity of the circuit. A common problem found in previous works is the use of quantum controlled gates from general unitaries, due to the diculty of physically implementing such gates. In this work, only CNOT controlled gates are used, resulting in circuits closer to physical reality. An evolutionary algorithm is proposed to decompose a unitary matrix into an equivalent sequence of quantum gates. The method is tested to nd decompositions of the Toffoli unitary, the two qubits quantum Fourier transform and the coin for the quantum walker. The results show that the algorithm is eficient in nding the solutions, and that the problem of premature convergence is avoided when populations are evolved in parallel