Quantum computers provide a fundamentally different model of computation that lends itself to naturally to a collection of problems that are expected to be infeasible or prohibitively inefficient to solve on classical computers. In my research, I focus on identifying such problems and formulating quantum computational solutions with the goal of reducing the computational resources required to solve the problems.
On one hand, I consider the efficacy of quantum optimisation methods for the purposes of solving optimisation problems that require evaluating a cost function across an exponentially large space of possible solutions, with the aim of speeding up the procedure of identifying optimal solutions.
On the other hand, I work with quantum machine learning techniques for the purposes of learning on datasets that involve high-dimensional or otherwise unknown underlying structures, a regime in which quantum computers are expected to work well owing to the fact that they inherently operate in high-dimensional "Hilbert spaces".
At UQ, my research focuses predominantly on the former, where I aim to utilise quantum optimisation methods for the purposes of solving integer linear programs or Hamiltonian simulation problems. However I am also greatly interested in identifying inherently quantum learning problems for which substantial empirical or provable improvements in learning performance over classical machine learning methods can be derived.
