🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

My book is (at last) out, just in time for Christmas! A blog post to celebrate and present it: francisbach.com/my-book-is-out/

First independent evaluation (by Han-Jia Ye et al.) of RealMLP in a large scale benchmark and.... we're first in rank among all methods 😎 Link at the end of the thread to the benchmark + RealMLP paper + how to try it in 5 mins

Learning rate schedules seem mysterious? Turns out that their behaviour can be described with a bound from *convex, nonsmooth* optimization. Short thread on our latest paper 🚇 arxiv.org/abs/2501.18965

Our new benchmark – AL4PDE – uses active learning to solve partial differential equations more efficiently with #neuralnetworks. It reduces data needs and lowers errors by up to 71% compared to random sampling. Learn how: neclab.eu/research-group…. #NECLabs #AI #MachineLearning

What if we have been doing early stopping wrong all along? When you break the validation loss into two terms, calibration and refinement you can make the simplest (efficient) trick to stop training in a smarter position Eugène Berta David Holzmüller Michael Jordan Francis Bach

Please join this timely workshop on Multi-scale Modeling for (Physical & Chemical) Sciences and Engineering problems. We propose LOGLO-FNO, an improvement to the popular FNO arch, specifically targeted towards modeling high frequencies in turbulence simulations. #ai4science #pde

🚨What is SOTA on tabular data, really? We are excited to announce 𝗧𝗮𝗯𝗔𝗿𝗲𝗻𝗮, a living benchmark for machine learning on IID tabular data with: 📊 an online leaderboard (submit!) 📑 carefully curated datasets 📈 strong tree-based, deep learning, and foundation models 🧵