Advances in Quantum Machine Learning

Abstract

Quantum machine learning is a broad term encompassing machine learning algorithms that are
either partly or entirely based on quantum information processing principles or entirely classical
machine learning algorithms used to solve quantum mechanical problems.
Here, we explore three realizations of quantum machine learning: a hybrid quantum-classical
generative adversarial network, hybrid quantum-classical variational algorithms, and finally
classical fermionic neural network Ansatz used in quantum Monte Carlo. In the first, we develop
the first implementation of a quantum-classical generative adversarial network, showing the
first, to our knowledge, known application to a hybrid quantum-classical model to a complex color
dataset. In the second, we demonstrate how classical machine learning algorithms, specifically
metalearning, can be advantageous to integrate with novel variational quantum algorithms for
optimization showing that metalearning methods have better performance than other commonly
used methods especially in the presence of parameter setting noise. Finally, in the third, we
extend existing work on variational Monte Carlo with fermionic neural network Ansatz by
improving the network design and further propagating the wave function with diffusion Monte
Carlo, achieving state-of-the-art performance on some small atomic systems (Be-Ne, C+).

Date of Award	28 Sept 2021
Original language	English
Awarding Institution	University of Bristol
Supervisor	I C G Campbell (Supervisor)