Simulating physics with Seurat's quantum cat

3 12/2012

Monday, 03 Dec. 2012, 15:00 - 16:00

Presenter: Peter Drummond - Swinburne University of Technology, Melbourne, Australia
Host: A. Zeilinger
Where: IQOQI Seminar Room, 2nd Floor, Boltzmanngasse 3, 1090 Vienna

Abstract: The increasing success of quantum information physics makes it timely to revisit the problem of probabilistic simulation of quantum paradoxes. Feynman once thought that such types of simulation would be impossible owing to Bell's theorem, unless one used some "hocus-pocus". We make the impossible possible using a synthetic universe with phase-space variables. We simulate Bell states, Schrodinger cats, and genuine multipartite Bell violations with GHZ states. This allows the demonstration of multipartite genuine Bell violations with up to 50 qubits - including the effect of ion-trap super-decoherence.

To achieve this, we divide up a quantum superposition into a myriad of probabilistic samples, like a pointillist Seurat painting. Also like a Seurat, once too close to the canvas, the illusion is gone. This is called the sampling error. We show that sampling error scales differently depending on the order of the correlation function and the representation method. Our work removes a limitation once thought to prevent probabilistic simulation of physics, and opens the door to new algorithms well-suited to modern GPU technology.

The detailed representations used will include positive-P, SU(2) Q-function, and the general Gaussian representations. We explain the "hocus-pocus" by mapping the probabilities of events in the synthetic universe, showing they do not satisfy the postulates of Bell. This is the reason why such phase-space simulations can take place probabilistically. In a nutshell, the computational universe events are not restricted to normal quantum eigenvalues, and instead behave more like weak measurements.

Practical simulation examples will also be treated, including down-conversion, first-principles theory of BEC formation, spontaneous vortices, and BEC quantum collisions leading to quantum correlated massive particles.