Postdoc @TUBerlin/@bifoldberlin. PhD in ML for quantum chem @ETH and @MIT

Switzerland
Joined May 2011
Normalize academics talking about their failures. I'll go first: We got rejected at Berghain on Saturday night.
31
93
35
2,097
309,180
I'll be at Pacifichem this week, hit me up if you wanna have a coffee! ☕
2
366
Dr. St❦fan Gugler retweeted
I'm in San Diego this week for NeurIPS, hit me up if you wanna have a coffee ☕
1
1
263
I'm in San Diego this week for NeurIPS, hit me up if you wanna have a coffee ☕
1
1
263
I'm in Oslo this week for WATOC, hit me up if you wanna have a coffee ☕
2
405
We have a new paper on diffusion!📄 Faster diffusion models with total variance/signal-to-noise ratio disentanglement! ⚡️ Our new work shows how to generate stable molecules in sometimes as little 8 steps and match EDM’s image quality with a uniform time grid. 🧵
2
13
1
33
4,222
First publication of my postdoc in Berlin is out on arXiv today! 🎉 ~~> We use diffusion probabilistic models to relax distorted molecules. No out-of-equilibrium data needed! A little 6-thread🧵
1
350
Dr. St❦fan Gugler retweeted
We are already on Day 3 of the workshop "Density Functional Theory and Artificial Intelligence learning from each other" in sunny CECAM-HQ. This afternoon, Fang Liu (@EmoryUniversity) & Michael Herbst (@EPFL_en) will present their talks.
3
5
45
2,193
Dr. St❦fan Gugler retweeted
If two models are more similar to each other than a third on ImageNet, will this hold for medical/ satellite images? Our preprint analyzes how vision model similarities generalize across datasets, the factors that influence them, and their link to downstream task behavior. 🧵1/7
1
14
5
57
14,323
Dr. St❦fan Gugler retweeted
🎉 Thrilled to announce that our paper MambaLRP has been accepted at #NeurIPS2024! Explain your Mamba models in a faithful, efficient, and effective way! 🐍 Huge thanks to @EberleOliver, Grégoire Montavon, and Klaus-Robert Müller @bifoldberlin, @ml_tuberlin , @GoogleDeepMind,
6
40
4
207
18,194
Equivariant daughter vs augmented son
who's going to win the election? | what's your paradigm: equivariance constraints or data augmentation?
63%Harris | equivariance
22%Trump | equivariance
7%Harris | augmentation
7%Trump | augmentation
27 votes • Final results
1
1
14
718
who's going to win the election? | what's your paradigm: equivariance constraints or data augmentation?
63%Harris | equivariance
22%Trump | equivariance
7%Harris | augmentation
7%Trump | augmentation
27 votes • Final results
2
1
4
1,237