This was an invited talk for the Waterloo Institute for Complexity and Innovation (WICI) seminar series. The talk was recorded and can be watched from WICI's website here.

Abstract:

Recent advances in Artificial Intelligence and Machine Learning (AI/ML) allow us to learn predictive models and control policies for larger, more complex systems than ever before. However, some important real world domains such as forest wildfire spread, flooding​ and medical imaging present a particular challenge. They contain spatially spreading processes(SSP) where some local features change over time based on proximity in space. This talk will present new approaches to​ learning for SSPs using Deep Learning and Reinforcement Learning such as learning a model of wildfire spread from satellite images as if the wildfire were an agent making decisions about where to move next. This approach could lead to learning models which are more interpretable and aid domain experts in creating rich, agent-based models based on data.