A common point raised by ML reviewers is that a method is too simple or is made of existing parts. But simplicity is a strength, not a weakness. People are much more likely to adopt simple methods, and simple ones are also typically more interpretable and intuitive. 1/2

And naturally, some hashtags! #NeurIPS2022 #NeurIPS NeurIPS Conference

This may revolutionize data science: we introduce TabPFN, a new tabular data classification method that takes 1 second & yields SOTA performance (better than hyperparameter-optimized gradient boosting in 1h). Current limits: up to 1k data points, 100 features, 10 classes. 🧵1/6

Joint Entropy Search neurips.cc/virtual/2022/p… This new acquisition function uses information about both location and value of the optimizer. It's intuitive, fast and could become a new default. Joint work with my great collaborators Carl Hvarfner & Luigi Nardi. Poster #502, Thu, 11am

In a couple of hours, we'll be presenting Joint Entropy Search neurips.cc/virtual/2022/p… NeurIPS Conference. Excited to discuss potential follow-ups for this powerhouse acquisition function! Many thanks to my collaborators Frank Hutter & Luigi Nardi. . Poster #502 @ 11am today.

Was my paper amazing? No. Did I expect one informed rebuttal reply? Yes. I think reviewing is fun, and try to be the reviewer I myself want to have. Do people genuinely dislike reviewing this much? #ICML2023

📢 Excited that I had the privilege of talking about "Harnessing new information in Bayesian optimization" Secondmind! 🎥👉youtu.be/-huaWITLyE8 Thanks to the organizers Hrvoje Stojic, Henry Moss, Victor Picheny, and collaborators Carl Hvarfner, Erik, Frank Hutter 🙌

I'm pleased to share that 3 of our research projects have been accepted for publication at NeurIPS Conference. 🚀 I'm looking forward to the contributions these projects can make to our community. 🔍 Here are the technical project summaries linkedin.com/posts/activity…

🎉 I’m delighted that my amazing team got 6 #AutoML papers accepted to #NeurIPS2023, including an oral on AutoML for Trustworthy AI! Other topics include #NAS, #BO, practical #HPO, LC extrapolation and #LLMs for data science. Congratulations to all my coauthors! Details below🧵1/

See also github.com/pytorch/botorc…. And x.com/luiginardi/sta….