My PhD research is focused on how we can design and develop new Artificial Intelligence (AI) devices which do not sacrifice performance or give rise to unwanted bias or unfairness but still respect the physical limits of their operating environment (in terms of power, size of chip and energy consumption). Within this research we are interested in the deployment of AI technology in smart sensors and wearable devices and how AI can be designed and implemented to solve a given problem.

The need for this research is due to the fact that with existing devices, bias and unfairness can be introduced (intentionally or unintentionally) at the detriment of individuals.  Many times, bias is introduced either through the algorithmic design of the device or by the data used to train an AI, or even a combination of both. Therefore, to tackle such issues of bias we need engineered solutions backed by thorough research and testing, especially now with AI becoming more ubiquitous in society.

Within my PhD research, the approach adopted aims to incorporate learnings from a variety of AI and computer science fields, in particular the methodology and techniques offered from neuromorphic computing. This will allow us to potentially benefit from the efficiency and performance gains it offers as well as provide us with the groundwork to develop our technology on and to focus on minimising bias and unfairness with our AI’s predictions.

We have identified a new AI architecture that we are currently applying to perform cardiac abnormality detection by processing the signals produced by an individual’s heart, something which can be crucial in the monitoring and early detection of heart disease. This technology allows for the possibility of deployment within a small and energy efficient wearable form factor while still remaining highly performant in real-time operation.

In my spare time and aside from my PhD, I have a great interest in using my knowledge and background in computer science to develop new AI technology which could be used for the betterment of others. I also have a great interest in gaming as well as mixed martial arts and I regularly train as a way to take a break from my studies. I am currently training for my 2^nd dan black belt in Ju-Jitsu and hope to attain this belt later this year.

Jarlath is a first year doctoral scholar on the Leverhulme Interdisciplinary Network on Algorithmic Solutions (LINAS) programme.   

He is one of the young entrepreneurs working in the world of digital selected for the Belfast Telegraph’s 30 under 30 for 2023, for excelling in their businesses, which vary from app creation, digital marketing and innovative online platforms. Jarlath is the founder and Chief Executive of Axonnr, an AI Robotics start-up company which specialises in using robotics to improve the lives of those with prosthetic limbs.