Estimating the degree of non-Markovianity using machine learning
Nenhuma Miniatura disponível
Data
2021-02-01
Autores
Fanchini, Felipe F. [UNESP]
Karpat, Göktuǧ
Rossatto, Daniel Z. [UNESP]
Norambuena, Ariel
Coto, Raúl
Título da Revista
ISSN da Revista
Título de Volume
Editor
Resumo
In the last few years, the application of machine learning methods has become increasingly relevant in different fields of physics. One of the most significant subjects in the theory of open quantum systems is the study of the characterization of non-Markovian memory effects that emerge dynamically throughout the time evolution of open systems as they interact with their surrounding environment. Here we consider two well-established quantifiers of the degree of memory effects, namely, the trace distance and the entanglement-based measures of non-Markovianity. We demonstrate that using machine learning techniques, in particular, support vector machine algorithms, it is possible to estimate the degree of non-Markovianity in two paradigmatic open system models with high precision. Our approach can be experimentally feasible to estimate the degree of non-Markovianity, since it requires a single or at most two rounds of state tomography.
Descrição
Palavras-chave
Como citar
Physical Review A, v. 103, n. 2, 2021.