Redes Neurais e Dinâmica Não-Markoviana

Abstract

The Neural Network is a powerful tool that is related to the application of "Machine Learning" in solving complex problems that require advanced computational devices, co- vering several applications extending to autonomous cars, biomedical and medical appli- cations such as discovery of new drugs, cure and early diagnosis of diseases, automatic generation of news and classification of protein structure, as well as pattern recognition such as facial recognition, product recommendation, prediction of future results as well as stock values from the stock exchange and climate change to fields of spacecraft engi- neering. In this sense, this work consists of the study of the lecture notes of "An introduction to neural networks for beginners"and the code implemented to train an RN that recog- nizes handwritten digits, as well as the study of non-Markovian dynamics of bipartite quantum systems in the presence of reservoirs and many-body interactions. The density matrix formalism and TCL are used to describe the quantum dynamics of open systems. Performing, from a theoretical point of view, what is called reservoir engineering, in order to obtain memory effects on the temporal evolution of the system, in which in the non- Markovian regime one seeks to perform entanglement calculations to obtain applications in quantum computing , inspired by the recent article by Banchi et al.