Causalidade, informação, causalidade da informação

Abstract

Quantum systems are able to show correlations between its parties that are stronger than correlations found in classical systems. Some of those strong correlations are called nonlocal, as for the classical correlations, they are calledlocal. Nonlocal correlations do not imply long-distance actions or superluminal communication, because they satisfy the nonsignaling condition. There are nonsignaling correlations stronger than quantum correlations because the quantum correlation set is a strictsubset of the nonsignaling correlation set. Quantum correlations are determined by how one obtains the probabilities of geting results conditioned to the measurement choices, considering the density operator describing the quantum state and the set of measurements. Moreover, a physical principle capable of justify quantum nonlocality is unkown. Because of that, an important question appeared:Would there be a physical principle which is able to explain the quantum nonlocality limits? Within the attempt to reach a positive answer for this question, some principles were presented in the last years. One of the most highlighted ones is the so-calledInformation Causality. This principle states that when one considers a system with two correlationedparties and the sending of a message from one party to the other containingd bits of information, the information gain within this process is limited tod bits. Quantum correlations satisfy the Information Causality principle. The best criteria known for the principle are able to discard most of the nonsignaling correlations that are stronger than quantum correlations, in the most simple scenarios. However, there still are supra-quantum correlations that obey such criteria. Recently, a new approach has been adopted for the achievement of a stronger criteriontorepresenttheInformationCausalityprinciple.Suchapproachisbasedonclassicaland quantum information theory, and causal inference theory, the latter being a powerful and interesting theory that relates the correlations with causal relations. This theory has found applications in many areas.This thesis presents the main elements of information theory, causal inference theory and nonlocality, aiming the analysis of the Information Causality principle and its strongest criterion. The main results obtained for the principle will be revised, and it will be presented a partial result obtained for a generalization of the more recent Information Causality criterion by considering scenarios in which multiple copies of the nonlocal resource are available.