Madsen Group. TUW

@olexandr One thing is the politics, but another is that this app simply sucks. Even if I was never particularly active, I used to discover new papers every week when the logo was a bird. Not anymore.

Academics retreating to 🦋 is a sign of weakness and denialism. The society is facing existential threats and possibly new dark ages. Going back to the ivory tower can hardly help us. Ironically dems in the US are making the same mistake:// #desperation #UnpopularOpinion

@dr_marossi Where are going? This place isn't what it used to be.

@HannesStaerk @n_gao96 Cool!

Monday with @n_gao96 in the reading group "Learning Equivariant Non-Local Electron Density Functionals" arxiv.org/abs/2410.07972 Join us on zoom at 9am PT / 12pm ET / 6pm CET: portal.valencelabs.com/logg

Funded by @FWF_at through @TACO_Oxides

Ralf's work on active learning facilitating the structural exploration of inhomogeneous surfaces is out in @digital_rsc. pubs.rsc.org/en/content/art…

funded by @FWF_at through @TACO_Oxides

Peter's work on generating structures using a GAN with a short-sighted critic is out in @MLSTjournal. LoGAN: local generative adversarial network for novel structure prediction doi.org/10.1088/2632-2…

Congratulation @esther_heid

@taylordsparks I'll definitely listen to this! "The making of the atomic bomb" is one of the most fascinating books I've read. Rhodes is very good at telling the side stories of all the characters involved, but I think even he missed Wilhelm's contribution.

You remember that scene in #Oppenheimer with the giant glass jars slowly filling with marbles to represent the uranium and plutonium needed for the Manhattan project? The project was saved by the Met Lab and Oak Ridge who provided the materials... Or was it???

Linux users rn

Congratulations @esther_heid !

@TimothyDuignan From: en.wikipedia.org/wiki/Coupled_c…

Many of the most important scientific advances have followed a simple recipe: Adopt a tool developed in another field to your own. The most beautiful demonstration of this is actually diffusion models which have done this three times already! Firstly tools and ideas from physics (statistical mechanics, molecular simulation etc.) were used to revolutionize image generation. Now that diffusion models have been massively improved and developed by computer scientists and engineers, they are being applied back to the field of molecular simulation, i.e., to generate plausible distributions of molecular structures where they are already having a huge impact. Finally they are being applied to design new medicines and materials. This is why interdisciplinary science is so crucially important. What are some other nice examples?

@COE_MECS It was a great event. Compact, lively and excellent talks. The poster session was really interesting as well. And we've hardly started yet.

Thanks to all participants for making this a great event ! © Matthias Heisler.

Looking forward to meeting old friends at the W2k workshop (indico.ictp.it/event/10467/) this week. I haven't been to the @ictpnews before, so I'm looking forward to that as well.

Happy to share our recent work! Pressure enhances electron disruption in phonon transport. @MadsenAg journals.aps.org/prl/abstract/1…

1/n: There are some academic papers that are so brilliantly and so accessibly written and so universal in scope that they transcend disciplines and stand as timeless testaments to both great thinking and great writing. Here's a short personal selection:

Nico's work on combining neural-network force fields and nested sampling for predictive phase diagrams is out as an editors' suggestion in @PhysRevMater. Very enjoyable collaboration with @lbpartay

How to verify the precision of DFT implementations via reproducible and universal workflows @icmabCSIC @psich_en @EPFL_en @nccr_marvel @max_center2 @aiidateam @madsenag nature.com/articles/s4225… Free to read 🔓

Welcome tu (!) the @tu_wien instagram.com/reel/Cx4x4NLMJ…