Throughout my academic career, I’ve been fortunate to contribute to a wide variety of research directions, including graph representation learning, computational biology, medical imaging, cross-modal learning, reinforcement learning, explainable AI, and deep learning on low-power/embedded systems. A detailed list of relevant publications is provided below.
Currently, my main research interests are on building machine learning systems aligned with algorithmic computation. This direction is likely to be important for novel reasoning breakthroughs, especially when extrapolating to inputs outside of their training data support. I’ve studied this direction from many angles, including but not limited to: neural algorithmic reasoning (NAR), algorithmic alignment, categorical deep learning (CDL), and length generalisation. I find it to be one of the final missing pieces on the path to generally intelligent systems, and am firmly committed to research in this space.
Lately, I have also been highly fascinated by the amazing things that happen when these methods interact synergistically with domain experts. The best examples of this are our Nature paper on guiding human intuition with AI, our Nature Communications paper on predictive and generative football analytics with TacticAI, and our technical report on AI-assisted competitive programming.
I would be happy to discuss any project idea at the intersection of the above areas; please check my contact page if you are interested!
