Benjamin Bloem-Reddy

University of British Columbia

Invariance in machine learning methods: What, why, and then?

Statistics Seminar

14th May 2021, 4:00 pm – 5:00 pm

Statistical and machine learning methods that incorporate appropriate symmetries have proven hugely successful, from exchangeable probability models and de Finetti's theorem in Bayesian statistics, to convolutional neural networks and their extensions in deep learning. I will give an overview of the what and the why, with an emphasis on recent work and some open problems, and look towards what might be next.

Comments are closed.