Skip to main content
SHARE
Publication

A PEPS Plugin for TNQVM

Publication Type
Conference Paper
Book Title
2023 IEEE International Conference on Quantum Computing and Engineering (QCE)
Publication Date
Page Numbers
382 to 384
Issue
.
Publisher Location
New Jersey, United States of America
Conference Name
IEEE International Conference on Quantum Computing and Engineering (QCE)
Conference Location
Bellevue, Washington, United States of America
Conference Sponsor
IEEE
Conference Date
-

This work introduces an extension to the Tensor Network Quantum Virtual Machine (TNQVM) tool, enhancing the existing stack of ExaScale Tensor Network (ExaTN), ExaScale Accelerator (XACC), and TNQVM. It features a new plugin that enables efficient simulation of a Projected Entangled Pair State (PEPS), a 2D tensor network. To improve simulation efficiency for PEPS, we have implemented the snake boundary contraction algorithm. By integrating this capability into the existing stack, we enhance the overall functionality and versatility of the framework. We tested this new PEPS topology for a simple GHZ bell-pair generation quantum circuit and saw that its runtime is very close to that of the MPS topology. We estimate that the real potential of the PEPS topology becomes discernible when quantum circuits with multidimensional entanglement are simulated using tensor networks. In such cases, 1D tensor networks fail to represent or contract them efficiently.