Tom Flynn

Hello!

My research focuses on scientific machine learning and optimization. I seek to advance the understanding of machine learning, including the relation between models, data, and training dynamics. Viewing this challenge from the lens of non-convex optimization, my work develops the analytical tools needed to understand and potentially improve the training of complex ML models.

Other areas I find interesting include (3D) computer vision, statistical change detection, kernel methods, and other mathy topics adjacent to ML. I'm open to collaborations - feel free to contact me.