Compressed Sensing Based on Tensor Network Machine Learning

Authors: Shao-Ming Fei*

DOI: 10.23880/psbj-16000174

Volume 5, Issue 1

Abstract

We introduce the scheme of compressed sensing based on tensor-network machine learning, which enables to compress and communicate information through the generative tensor-network states. The state \(\left| \Psi \right\rangle\) is first obtained by unsupervised learning of tensor network, which characterizes the set of training images. With  \(\left| \Psi \right\rangle\) and a small amount of pixels from a specific image that is to be sent, one can obtain a projected state  \(\left| \Psi \right\rangle\), from which the whole sent image can be reconstructed. The key problem of selecting pixels from a specific image is solved by investigating the entanglement in the state\(\left| \Psi \right\rangle\).

Keywords: Compressed sensing; Tensor network; Machine learning

View PDF