I gave a talk at the TU Berlin.

Thanks a lot to the organizer for inviting me!

You can find the slides here.

Here is the abstract:

Tabular Foundational Models: An Overview:

Tabular foundational models (TFMs) have made great progress recently. While they are still limited to small/medium datasets due to the quadratic complexity of self-attention, they now dominate other methods for small/medium datasets. In this talk, I will briefly explain how TFMs work before discussing some of their limitations, in particular their lack of “built-in” symmetries which cause instabilities in high-dimensional classification settings. I will then present an application of TFMs to time-series forecasting, showing the potential impact of these approaches. Finally, I will conclude by presenting TabArena, a recent benchmark developed for tabular models.