Abstract for the talk on 19.10.2023 (17:00 h)
Math Machine Learning seminar MPI MIS + UCLADaniel Murfet (University of Melbourne)
Phase transitions in learning machines
19.10.2023, 17:00 h, only Live Stream
I will introduce the idea of phases and phase transitions of the Bayesian posterior in the setting of singular learning theory, and discuss how a simple auto-encoder model introduced by Anthropic in their research on neural network interpretability displays a rich set of phase transitions in both the posterior and over the course of training. I’ll explain a research program we term “developmental” interpretability that is aiming to use phase transitions as the basic primitive for understanding the internal structure of computation in neural networks.
If you want to participate in this Live Stream please register using this special form. The (Zoom) link for the Live Stream will be sent to your email address one day before the seminar.