Our new work has appeared on arXiv: arxiv.org/abs/2408.05215, which addresses a challenge in #Machinelearning interatomic potentials by introducing a #physics-informed, #weakly supervised approach.

Tomorrow I'll be giving a talk on Generalization Theory of Linearized Neural Networks at the MML seminar: mis.mpg.de/events/event/g… The (virtual) talk will be live-streamed (5pm CEST = 11am EST = 8am PST) and a recording will be uploaded later.

"Mindless molecules" are making a comeback! 🚀 We’ve launched a fresh SotA Python implementation inspired by the original 2009 paper. In our minds, essential molecule data points for building truly transferable SQM or MLFF models! 🔬📊 github.com/grimme-lab/Min…

The video and slides of my talk are online (link is in the quoted tweet). 📽️

Did you ever wonder how to combine electrochemistry and acid/base chemistry 🧪💻🔋? In doi.org/10.1002/celc.2… we present a straight-forward way. Can be based on experimental or computational data. With Patrick M. Becker, Heinze_Lab, Sarkar Lab. Thanks to Studienstiftung!

#NECLabs is accelerating #chemicalsciences with #neuralnetworks built using higher-rank irreducible Cartesian tensors – our new machine-learned interatomic potential! Our model matches DFT accuracy at a fraction of its computational cost. Learn how: neclab.eu/research-group….

To introduce you to ATLAS, we have put together an advent calendar. Each day we will present a sub-project. Look forward to what awaits you each day on Instagram and X. 🕯️Our advent calendar starts with a video from Prof. Kästner explaining what ATLAS's Project Area A covers.

I'll be at #NeurIPS2024 next week to present this paper (Thu afternoon) as well as a workshop paper on active learning for neural PDE solvers. Let me know if you'd like to chat about tabular data, uncertainty, active learning, etc.!

Experimental characterization of single-site platinum catalysts is tricky. Esteslab and Aaron Rossini used solid-state NMR for 1H, 31P, and 195Pt - we contributed DFT results 🧪💻to understand immobilized platinum hydrides. CRC1333 See: Chemical Science doi.org/10.1039/D4SC06…

🚨Can we go beyond state-of-the-art message-passing models based on spherical tensors such as #MACE and #NequIP? Our #NeurIPS2024 paper explores higher-rank irreducible Cartesian tensors to design equivariant #MLIPs. Paper: arxiv.org/abs/2405.14253 Code: github.com/nec-research/i…

Join us today at #NeurIPS2024! Higher-Rank Irreducible Cartesian Tensors for Equivariant Message Passing 🗓️ When: Wed, Dec 11, 11 a.m. – 2 p.m. PST 📍 Where: East Exhibit Hall A-C, Poster #4107 #MachineLearning #InteratomicPotentials #Equivariance #GraphNeuralNetworks

"What Did I Do Wrong? Quantifying LLMs' Sensitivity and Consistency to Prompt Engineering' will appear in NAACL. Thanks to Federico and all co-authors for the fun work (Link👇) You can guess what is a regular source of frustration in the modern software development process...

* Graph Learning Will Lose Relevance Due To Poor Benchmarks * Maya Bechler-Speicher, Ben Finkelshtein, Fabrizio Frasca, Luis Muller, et al. Inc. Michael Bronstein Bryan Perozzi Michael Galkin Christopher Morris Wow we should be discussing / acting on this!

Watch NEC Senior Research Engineer Dr. Daniel Oñoro Rubio explain how our materials informatics platform can help the materials industry speed up #materialdiscovery & development, ensure regulatory compliance & uncover business opportunities. youtu.be/1YgToKtBnHQ #NECLabs #AI

Estimating Absolute Protein–Protein Binding Free Energies by a Super Learner Model #MachineLearning pubs.acs.org/doi/10.1021/ac… Matheus Ferraz #JCIM Vol65 Issue5 #compchem

If you are at ICLR and interested in ways to make denoising diffusion more efficient, please come to Vinh’s oral talk tomorrow at 11:30 in oral session 1C. It also involves backprop through ODE solvers and constrained learning.

📣 Our spicy ICML 2025 position paper: “Graph Learning Will Lose Relevance Due To Poor Benchmarks”. Graph learning is less trendy in the ML world than it was in 2020-2022. We believe the problem is in poor benchmarks that hold the field back - and suggest ways to fix it! 🧵1/10

Workshop: Imagining Interstellar Ice: Across Art and Science Can artists and scientists explore interstellar ice together? We’re inviting early-career artists and scientists to a one-of-a-kind workshop +