IDA Machine Learning Seminars - Fall 2024

Thursday, September 26 at 14:15, 2024

Modelling and generating data via deep probabilistic representations

Thomas Schön, Department of Information Technology, Uppsala University

Abstract: One of the key lessons to take away from contemporary machine learning is that flexible models often offer the best predictive performance. This has implications in many situations. In this talk I will try to make this concrete by looking at a few constructions that we are working with. I will start with a (classical) classification task from ECG interpretation and then continue to the more under-researched area of how to formulate and solve regression problems using deep learning. There are currently several different approaches used for deep regression and there is still room for innovation. I will illustrate this landscape in general and introduce some of our developments consisting of a deep regression method which has a clear probabilistic interpretation. When it comes to generative models I will also share some insights related to diffusion models, in particular related to its use for image restoration. Besides sharing some of our findings for this particular problem I will also point out some more general aspects we came to realize in working on this.

Location: Alan Turing