Quantum mechanics has many nonclassical features, including uncertainty relations, tunneling, and entanglement. One of the most remarkable features is the prediction of nonclassical correlations between two distant quantum systems, that cannot be explained by any local and deterministic theory in which experimenters can freely choose what to measure.
Such correlations have been experimentally verified (via the violation of so-called Bell inequalities), and underly protocols for secure quantum cryptography and quantum random number generation. Hence, there is great interest in their interpretation.
The only way of giving a local and deterministic interpretation of these strong quantum correlations is via models that allow a statistical dependence between the choice of measurement settings and hidden system variables. The efficiency of such models can be measured in terms of the amount of information residing in this dependence. This project will explore maximally efficient models, in the sense of requiring minimal information, for the case of entangled quantum bits.
Basic knowledge of the quantum formalism is required, including the description of quantum bits (qubits), as well as the ability to manipulate classical joint probabilities.