My interests are in computational ecology. My group works in two areas: AI for environmental monitoring and mathematical / computational models for animal group behaviour.

We develop AI methods and systems for monitoring animal activity as the basis for biodiversity monitoring. This branch of my work involves deep learning, classical ML techniques, and Edge-AI for both audio and video data as well as statistical modelling of monitoring outcomes. 

We develop mathematical models to explain how organisms interact with their environment, including the impact our changing environment has on their lives.  Most of my work is centred on the collective behaviour of social insects, such as bees and ants, in the hope that a deeper understanding of their behaviour will allow us to better protect them and the important ecosystem services they provide. Insect societies are entirely self-organised, without any central leader or master plan. How such decentralised “super-organisms" plan and coordinate their actions is still a largely open question. I work closely with experimental biologists to decipher the mechanisms behind this. We pursue multi-model approaches, where the behavioural complexity is initially captured in individual-based simulations and then distilled into (often analytically treatable) mathematical models of the core functionality. My work uses and extends a range of mathematical and computational techniques, including reinforcement learning, evolutionary game theory, reaction-advection diffusion, and stochastic event analysis to explain this. 

For more info see my personal homepage.