🤖 A big thank you to the Pioneer Centre for AI (Pioneer Centre for AI) for your invaluable support and sponsorship of the Nordic #ProbAI 2024.

📊 Nordic #ProbAI 2024 is grateful for the generous support from the Danish Data Science Academy (Danish Data Science Academy). Your efforts in building a flourishing data science ecosystem are significant.

🍻 A heartfelt thank you to the Carlsberg Foundation (Carlsbergfondet) for your generous sponsorship of Nordic #ProbAI 2024. Your commitment to supporting scientific research, education, culture and art is greatly valued.

👏 A big thank you to the teams from Aalborg Universitet, @norwegianopenal and NTNU/NTNUnorway for supporting Nordic #ProbAI 2024.

Do you also wonder what Bayesian Workflows are? Andrew Johnson (Aalto University) dedicates a session to it and provides tutorials on the topic at #ProbAI.

Wouter Boomsma (Wouter Boomsma) introduced us to Bayesian optimization, teaching how to learn better functions for protein engineering #ProbAI

It’s time for Helge Langseth & Thomas Dyhre Nielsen with variational inference and optimization, including many tutorials, to get hands-on practice on this foundational topic #ProbAI

The 3rd day opens with Deep Generative Models by Jes Frellsen, including Normalizing Flows, DLVMs, and more #ProbAI

What a nice finish to the Deep Generative Models day with Ole Winther on "SDEs are all you need for latent variable models", followed by a poster session for more discussions and to connect with others #ProbAI

We are deeply grateful to all our sponsors Pioneer Centre for AI, Danish Data Science Academy, and Carlsbergfondet, and members from Aalborg Universitet, @norwegianopenal and NTNU/NTNUnorway for making this edition possible. 🇩🇰

arxiv.org/pdf/2406.13661 A very nice survey on generative learning from a statistical physics perspective, credits to Davide Carbone !

⚡️ Excited to share that I am starting an AI+Education company called Eureka Labs. The announcement: --- We are Eureka Labs and we are building a new kind of school that is AI native. How can we approach an ideal experience for learning something new? For example, in the case

The Cold Posterior Effect Indicates Underfitting, and Cold Posteriors Represent a Fully Bayesian ... Yi-Shan Wu, Yijie Zhang, Luis A. Ortega, Andres R Masegosa. Action editor: Vincent Fortuin. openreview.net/forum?id=GZORX… #posteriors #underfitting #posterior

Excited to publish a Python package that turns Andrej Karpathy's "A Recipe for Training Neural Networks" into easy-to-use diagnostics code! 🔧 No more randomly poking around in your custom PyTorch DNN to debug it. Get simple diagnostics for your neural nets 🫶 #PyTorch 1/

arxiv.org/abs/2010.06610 is such an insane paper and idea 🤯

Terence Tao, a superstar in maths, discusses opportunities at the intersection of AI & maths. youtu.be/e049IoFBnLA?si… Takeaways: - ML and LLMs unlock new areas of maths previously challenging for humans. - Useful when problems or proofs exceed human scales. - ML models make

BREAKING : in the latest large scale study XGBoost comes last in the top 10. As shown by all recent large scale studies CatBoost rules the roost. It's time to set the record straight: XGBoost might get a lot of attention on social media, but when it comes to handling tabular

This is one of my favorite papers to assign for incoming graduate students. And, re-reading it for this morning's class gave me some reminders I needed myself. ✨ journals.biologists.com/jcs/article/12…

Large Language Model Agents is the next frontier. Really excited to announce our Berkeley course on LLM Agents, also available for anyone to join as a MOOC, starting Sep 9 (Mon) 3pm PT! 📢 Sign up & join us: llmagents-learning.org

The Llama 3 paper is a must-read for anyone in AI and CS. It’s an absolutely accurate and authoritative take on what it takes to build a leading LLM, the tech behind ChatGPT, Gemini, Copilot, and others. The AI part might seem small in comparison to the gargantuan work on *data*