Quantum Institute: Visitor Schedule

The Quantum Lunch is regularly held on Thursdays in the Theoretical Division Conference Room, TA-3, Building 123, Room 121.

The organizing committee includes Ryan O. Behunin (T-4 & CNLS), Malcolm Boshier (P-21), Adolfo del Campo (T-4 & CNLS), Michael Di Rosa (C-PCS), Ivar Martin (T-4), Changhyun Ryu (P-21), Rolando D. Somma (T-4), Christopher Ticknor (T-1), and Wojciech H. Zurek (T-4).

For more information, or to nominate a speaker, contact Adolfo del Campo.

To add your name to the Quantum Lunch email list, contact Ellie Vigil.

Tuesday, November 6, 2012
3:00 PM - 4:00 PM

Speaker: Fernando Cucchietti

(The Barcelona Supercomputer Center (BSC))

Technical Host: Eddy Timmermans

TOPIC: Can we trust quantum simulators?

Abstract
Quantum simulators can compute the behavior of interesting quantum systems exponentially faster than current computers. Because they are tailored to imitate specific models, it is widely believed that simulators are much easier to implement than universal quantum computers, which has fueled an explosive growth of research in the field. However, unlike classical and quantum computers, quantum simulators have no known way of self-correcting errors in a fault-tolerant manner: the validity of their results can thus not be assured. Even more, the robustness of quantum simulators in the presence of noise and at finite temperature is poorly understood to date. Therefore, we do not know if, in realistic experimental situations, quantum simulators can be truly more powerful than traditional computers. I will start by proposing a formal definition of a useful quantum simulator (basically, that it must satisfy four conditions: relevance, controllability, reliability, and efficiency). Then, I will perform an analysis of reliability and efficiency in the presence of static imperfections. I will illustrate these concerns with numerical simulations of a paradigmatic example: a disordered quantum spin chain governed by the Ising model in a transverse magnetic field. I will end with some speculation on what research is needed (theoretical and experimental) to achieve a level of confidence at least partially similar to that seen in computers with built-in error correction.