Podcast, Machine Learning Street Talk (MLST)
How Deep Learning Finally Cracked Messy Tables - Frank Hutter
About
Frank Hutter, co-founder of Prior Labs, talks about TabPFN, a tabular foundation model that makes predictions in a single forward pass, and the research behind it.
TabPFN is pre-trained on synthetic datasets drawn from a prior over structural causal models, rather than on real data. At prediction time it takes the whole training table as context and outputs an approximation of the Bayesian posterior predictive distribution, without per-dataset training or hyperparameter search. Frank explains how this grew out of his earlier work on AutoML and neural architecture search, how the priors are built and revised, and why tabular data was hard for deep learning for so long.
The conversation also covers the TabArena benchmark, how the architecture changed from TabPFN v1 to v3, scaling to larger tables, using the model with coding agents, test-time compute, Google's TabFM, causal inference and interventions, and relational data. At the end, a short update Frank recorded after the interview covers the TabPFN-3.5 release.
Topics
- TabPFN, search World Wide
- AutoML, search World Wide
- tabular foundation model, broader topic: foundation model
- synthetic datasets
- structural causal models, search World Wide
- Bayesian posterior predictive distribution, search World Wide
- neural architecture search, search World Wide
- TabArena, search World Wide
- test-time compute, search World Wide
- relational data
Details
Show
Machine Learning Street Talk (MLST)6 episodes on World Wide · Machine Learning Street Talk (MLST)