Constrained quantum learning
Using machine learning to train a Gaussian quantum circuit with PNRs to produce cubic phase resource states with high fidelity and probability.
This repository contains the source code used to produce the results presented in "Near-deterministic production of universal quantum photonic gates enhanced by machine learning" arXiv:1809.04680.
The following two scripts perform a constrained variational quantum circuit optimization, using both a global search (basin hopping) and a local search (BFGS optimization) to maximize the fidelity (and probability of generating) the cubic phase resource state in the last mode.
two_mode.py: a Python script to generate the results of the two-mode gadget architecture presented in the paper. Here, a two mode squeezed displaced state is incident on a beamsplitter, with the first mode measured by a photon-number resolving detector.
three_mode.py: a Python script to generate the results of the three-mode gadget architecture presented in the paper. Here, a three mode squeezed displaced state is incident on an interferometer consisting of three beamsplitters, with the first and second modes measured by photon-number resolving detectors.
To construct and optimize the constrained variational quantum circuits, these scripts use the Fock backend of Strawberry Fields. In addition, SciPy is required for use of the global Basin Hopping optimization method, as well as the local BFGS optimization method.
Krishna Kumar Sabapathy, Haoyu Qi, Josh Izaac, and Christian Weedbrook.
If you are doing any research using this source code and Strawberry Fields, please cite the following two papers:
Krishna Kumar Sabapathy, Haoyu Qi, Josh Izaac, and Christian Weedbrook. Near-deterministic production of universal quantum photonic gates enhanced by machine learning. arXiv, 2018. arXiv:1809.04680
Nathan Killoran, Josh Izaac, Nicolás Quesada, Ville Bergholm, Matthew Amy, and Christian Weedbrook. Strawberry Fields: A Software Platform for Photonic Quantum Computing. arXiv, 2018. arXiv:1804.03159
This source code is free and open source, released under the Apache License, Version 2.0.