What is it about?

We show, within the language of tensor network states, how information about low lying excited states can be extracted from just the ground state of a low dimensional strongly correlated quantum many body system with local interactions. More specifically, we show how the eigenspectrum of the tensor network transfer matrix, a central local object in tensor network techniques, contains this information and can be extracted without any prior knowledge of the underlying Hamiltonian.

Featured Image

Why is it important?

While obtaining ground states of strongly correlated many body systems already forms a formidable task by itself, extracting information about excited states is an even harder task. Our approach enables easy first access to information about low lying excited states, especially for quantum systems in 2 or higher dimensions, for which no competitive algorithms for obtaining excited states exist.

Perspectives

I think it is amazing how much information is already encoded in a local object describing the ground state of a strongly correlated quantum many body system. The fact that we can extract information about excited states from just the ground states demonstrates again how special strongly correlated systems with local interactions are. I find it particularly intriguing, how easy one can extract a first estimate of excitation energies in two dimensional systems (also with topological order) if the ground state is known in terms of a tensor network (such as PEPS).

Mr Valentin Zauner
Universitat Wien

Read the Original

This page is a summary of: Transfer matrices and excitations with matrix product states, New Journal of Physics, May 2015, Institute of Physics Publishing,
DOI: 10.1088/1367-2630/17/5/053002.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page