Machine Learning Street Talk Video posted an update
How Deep Learning Finally Cracked Messy Tables – Frank Hutter
Why it mattersFrank Hutter discusses how TabPFN uses synthetic datasets drawn from structural causal model priors and treats a training table as context to approximate Bayesian predictions in one forward pass. The conversation also covers scaling, test-time compute, causal inference and the model’s development from v1 to v3.
Discuss: How does TabPFN’s use of synthetic-data priors and whole-table context change the trade-offs of making Bayesian predictions in one forward pass?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.