Dresden 2020 – scientific programme
The DPG Spring Meeting in Dresden had to be cancelled! Read more ...
Parts | Days | Selection | Search | Updates | Downloads | Help
DY: Fachverband Dynamik und Statistische Physik
DY 53: Poster: Quantum Dynamics, Chaos and Information; Many Body Systems
DY 53.20: Poster
Thursday, March 19, 2020, 15:00–18:00, P1A
Performance of locally purified TN states as Ansatz-class for physical systems — •Lennart Bittel1, Albert H. Werner2, and Martin Kliesch1 — 1Heinrich Heine Universität, Düsseldorf, Germany — 2University of Copenhagen, Copenhagen, Denmark
Tensor network models have proven to be a useful tool for the simulation of interacting quantum many-body systems and have also been extended from closed to open quantum systems. Matrix product density operators (MPDO) provide a common ansatz class to simulate quantum states under nearest neighbour Lindblad dynamics. Alternatively, locally purified tensor networks can be used to avoid positivity issues and to to obtain trace norm control for the simulation errors. However, It is known that some states can be approximated less efficiently compared to MPDO. Moreover, purifications of quantum states have a unitary gauge freedom, which can lead to significant errors in simulation if not appropriately fixed. In this work, we investigate use cases and develop algorithms for open tensor networks like state approximation, Lindblad dynamics and finding the respective equilibrium state and show that purified networks can outperform MPDO for relevant physical systems.