The third talk within Seminar series in Statistics and Data Science of this semester, will take place at Erling Svedrups plass (Niels Henrik Abel Hus, 8th floor), on MONDAY ( NOTE, A SPECIAL OCCASION), 18.10.2021, 14:15. The talk will be given by Håvard Rue, a professor at King Abdullah University of Science and Technology.

Access to the seminar area physically at this point is not planned to be restricted. Yet, it will be possible to follow the talk on Zoom, using this link: 

https://uio.zoom.us/j/66205573085?pwd=T3lnMC85N3FGSmFxeXFvcHFnd2loUT09

Title:
Two applications of the variational form of Bayes theorem

Abstract:
In this talk I will discuss the variational form of Bayes theorem by Zellner (1988). This result is the rationale behind the variational (approximate) inference scheme, although it is not always that clear in modern presentations. I will discuss two applications of this results. First, I will show how to do a low-rank mean correction within the INLA framework (with amazing results), which is essential for the next generation of the R-INLA software currently in development. In the second one, I will introduce the Bayesian learning rule, which unify many machine-learning algorithms from fields such as optimization, deep learning, and graphical models. This includes classical algorithms such as ridge regression, Newton’s method, and Kalman filter, as well as modern deep-learning algorithms such as stochastic-gradient descent, RMSprop, and Dropout.

The first part of the talk is based on our recent research at KAUST, while the second part is based upon arxiv.org/abs/2107.04562 with Dr. Mohammad Emtiyaz Khan, RIKEN Center for AI Project, Tokyo.

Best regards,
Sven Ove Samuelsen & Aliaksandr Hubin