A quantum neural network computes its own relative phase

TL;DR

This paper introduces a quantum neural network that can autonomously determine its own relative phase, a key factor in quantum entanglement and computation, marking progress in quantum state characterization.

Contribution

It demonstrates a novel approach where a two-qubit quantum neural network is trained to compute and output its own relative phase, advancing quantum state analysis methods.

Findings

The quantum neural network successfully computes its own relative phase.

This method provides a new way to characterize quantum states.

It highlights potential for autonomous quantum state analysis.

Abstract

Complete characterization of the state of a quantum system made up of subsystems requires determination of relative phase, because of interference effects between the subsystems. For a system of qubits used as a quantum computer this is especially vital, because the entanglement, which is the basis for the quantum advantage in computing, depends intricately on phase. We present here a first step towards that determination, in which we use a two-qubit quantum system as a quantum neural network, which is trained to compute and output its own relative phase.