Michael Catchen

Supervised by: Timothée Poisot
Université de Montréal

Deep learning for species distribution modeling and monitoring

My postdoc project will apply deep learning methods to answer one of the most fundamental questions about life on Earth: where are species? Species Distribution Models (SDMs) are widely used by ecologists to make decisions about biodiversity conservation, agricultural pest control, and the management of invasive species. My work will focus on several challenges in this domain, including (1) application of modern tools from computer vision to SDMs, (2) methods for species representation and transfer learning to fill in gaps for data deficient species, and (3) semi-supervised methods to optimize the spatial design biodiversity monitoring programs.