Optical Quantum Mixed-State Reconstruction With Multiple Deep Learning Approaches

TL;DR

This paper introduces two neural network-based methods for quantum state tomography, improving accuracy for pure and mixed states by leveraging class information, representing a significant advancement in quantum state reconstruction techniques.

Contribution

The paper presents two novel neural network approaches that enhance quantum state tomography for pure and mixed states, utilizing class information for improved performance.

Findings

Achieved state-of-the-art accuracy in quantum state tomography.

Effectively reconstructed both pure and mixed quantum states.

Leveraged class information to improve neural network performance.

Abstract

Quantum state tomography is a crucial technique for characterizing the state of a quantum system, which is essential for many applications in quantum technologies. In recent years, there has been growing interest in leveraging neural networks to enhance the efficiency and accuracy of quantum state tomography. However, versatile methods that are broadly applicable across diverse reconstruction scenarios remain relatively underexplored. In this paper, we present two neural network-based reconstruction approaches for both pure and mixed quantum state tomography: Restricted Feature Based Neural Network and Mixed States Neural Network. By leveraging class information during reconstruction, we are able to achieve state-of-the-art performance of tomography for both pure and mixed quantum states.