The problem of computing the quantum dynamical entropy introduced by Alicki and Fannes requires the trace of the operator function $F(\Omega) = - \Omega \log \Omega$, where $\Omega$ is a non-negative, Hermitean operator. Physical significance demands that this operator be a matrix of large order. We study its properties and we derive efficient algorithms to solve this problem, also implementable on parallel machines with distributed memory. We rely on a Lanczos technique for large matrix computations developed by Gene Golub.
Quantum dynamical entropy and an algorithm by Gene Golub
MANTICA, GIORGIO DOMENICO PIO
Primo
2008-01-01
Abstract
The problem of computing the quantum dynamical entropy introduced by Alicki and Fannes requires the trace of the operator function $F(\Omega) = - \Omega \log \Omega$, where $\Omega$ is a non-negative, Hermitean operator. Physical significance demands that this operator be a matrix of large order. We study its properties and we derive efficient algorithms to solve this problem, also implementable on parallel machines with distributed memory. We rely on a Lanczos technique for large matrix computations developed by Gene Golub.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.