Lab Home | Phone | Search
Center for Nonlinear Studies  Center for Nonlinear Studies
 Colloquia Archive 
 Postdoc Seminars Archive 
 Quantum Lunch 
 Quantum Lunch Archive 
 CMS Colloquia 
 Q-Mat Seminars 
 Q-Mat Seminars Archive 
 Kac Lectures 
 Dist. Quant. Lecture 
 Ulam Scholar 
 CNLS Fellowship Application 
 Summer Research 
 Past Visitors 
 History of CNLS 
 Maps, Directions 
 CNLS Office 
Thursday, December 08, 2016
12:30 PM - 1:30 PM
T-DO Conference Room (03-123-121)

Quantum Lunch

To protect and to increase quantum coherence

Lorenzo Campos Venuti

I will discuss two questions related to quantum coherence. The first question is whether it is possible to perform quantum information processing in presence of strong dissipation. I will consider a strongly dissipative system admitting a non-trivial manifold of steady states. I will show how one can enact quantum logical gates inside this steady-state manifold by adding a weak, time-rescaled, Hamiltonian term into the system’s Liouvillian. The effective dynamics is governed by a projected Hamiltonian which results from the interplay between the weak unitary control and the fast relaxation process. The leakage outside the steady-state manifold entailed by the Hamiltonian term is suppressed by an environment-induced symmetrization of the dynamics. Given sufficient resources this approach allows for universal quantum computation and even completely dissipative-generated quantum gates. In the second part I will discuss adiabatic quantum computation in presence of noise and in particular examine whether a speedup can survive at finite temperature. I will introduce a framework where the adiabatic preparation time can naturally be compared to the relaxation time of a classical Markov chain. Using known bounds for the mixing time and for the open-system adiabatic theorem, one is led to a rather astonishing paradox. Namely relaxation seems always faster than adiabatic preparation. In systems describing thermalization I resolve the paradox by estimating an adiabatic bound in the low temperature limit. In this seeting, systems admitting a speedup at zero-temperature will still show a speedup for sufficiently low temperature.

Host: Rolando Somma