Neural-network quantum state tomography

TL;DR

This paper demonstrates that neural networks can be effectively used for quantum state tomography, ensuring valid quantum states and enabling advanced deep learning methods for noisy quantum state reconstruction.

Contribution

It introduces a method to implement the positivity constraint in neural networks for quantum state tomography, adaptable to any neural network architecture.

Findings

Neural networks can be trained to produce valid quantum states.

The method is compatible with standard neural network architectures.

It enables quantum state reconstruction under various noise conditions.

Abstract

We revisit the application of neural networks techniques to quantum state tomography. We confirm that the positivity constraint can be successfully implemented with trained networks that convert outputs from standard feed-forward neural networks to valid descriptions of quantum states. Any standard neural-network architecture can be adapted with our method. Our results open possibilities to use state-of-the-art deep-learning methods for quantum state reconstruction under various types of noise.