qNetVO: The Quantum Network Variational Optimizer

Simulate and optimize quantum communication networks using quantum computers.

Features

QNetVO simulates quantum communication networks on differentiable quantum cicuits. The cicuit parameters are optimized with respect to a cost function using automatic differentiation and gradient descent. QNetVO is powered by PennyLane, an open-source framework for cross-platform quantum machine learning.

Simulating Quantum Communication Networks:

• Construct complex quantum network ansatzes from generic quantum circuit compenents.
• Simulate the quantum network on a quantum computer or classical simulator.

Optimizing Quantum Communication Networks:

• Use our library of network-oriented cost functions or create your own.
• Gradient descent methods for tuning quantum network ansatz settings to minimize the cost.

Quick Start

Install qNetVO:

$ pip install qnetvo

Install PennyLane:

$ pip install pennylane==0.33

Import packages:

import pennylane as qml
import qnetvo as qnet

Note

For optimal use, qNetVO should be used with PennyLane. QNetVO is currently compatible with PennyLane v0.33.

Contributing

We welcome outside contributions to qNetVO. Please see the Contributing page for details and a development guide.

How to Cite

See CITATION.bib for a BibTex reference to qNetVO.

License

QNetVO is free and open-source. The software is released under the Apache License, Version 2.0. See LICENSE for details and NOTICE for copyright information.

Acknowledgments

We thank Xanadu, the UIUC Physics Department, and the Quantum Information Science and Engineering Network (QISE-Net) for their support of qNetVO. Work funded by NSF award DMR-1747426.