Loss is ubiquitous in natural systems, due to interactions with their environment. For information processing, it is a way to lose entropy, which is arguably what you need to do to extract only the salient features from data. Instead of seeing loss as an imperfection to be engineered around, can we harness it? For example, the natural properties of photons include that they aren’t conserved – one photon can be split into two lower energy photons. As part of a funded research project, we are currently developing ways to compute that use photon loss -and gain - as features, not bugs.

This PhD project will start from these novel ways to compute and develop the potential applications they are most suited for, devising test algorithms that can be run on existing hardware, such as the photonic quantum computer hosted by the National Quantum Computing Centre. There are many possible directions the research could take, and there will be freedom to explore multiple promising avenues depending on your interests.