Professor for Computational Chemistry of Nanomaterials @uni_kassel. Married, two children & has a dog.
Kassel, DE
Joined June 2011
- Tweets110
- Following620
- Followers962
- Likes165
Guido von Rudorff retweeted
Check out this CECAM workshop about machine learning + enhanced sampling + nuclear quantum effects! I am looking forward to learn something new!
Few weeks for application to this exciting CECAM workshop in Lausanne 29/11-01/12 2023.
As all cecam workshop there are no registration fees and we have a great "parterre" of invited speakers!
@cecamEvents @KarenPalacioR
cecam.org/workshop-details/1…
Guido von Rudorff retweeted
1/n Excited to share our preprint (@ferchault , Konstantin @KKarandashev): "Understanding Representations by Exploring Galaxies in Chemical Space"! 🌌 A Monte Carlo approach to probe chemical feature spaces without exhaustive enumeration or ML training.🚀
arxiv.org/abs/2309.09194
Guido von Rudorff retweeted
this paper's nuts. for sentence classification on out-of-domain datasets, all neural (Transformer or not) approaches lose to good old kNN on representations generated by.... gzip aclanthology.org/2023.findin…
Guido von Rudorff retweeted
Retweets are welcome and of course applications, especially from female students! :)
If you know someone looking for a PhD to develop #ML methods for chemical applications, preferably photochemistry, spread the word! tinyurl.com/2p935dfv
This quoted post is unavailable.
Guido von Rudorff retweeted
I couldn't be happier to attend our cecam workshop on #ML in beautiful Vienna, co-hosted by @mms_boku
cecam.org/workshop-details/1…
Good to see so many amazing people again & having 2 students from @UniLeipzig here @ferchault @BettinaLier @PePoliak @max_jr @Brigitta_B_ @BarrettRhyan
Guido von Rudorff retweeted
Nature's recent editorial, “For Chemists, the AI Revolution has yet to Happen,” argues that “Machine-learning systems in chemistry need accurate & accessible training data. Until they get it, they won’t achieve their potential.” See if you agree.
nature.com/articles/d41586-0…
Guido von Rudorff retweeted
Private data can be used securely for training and prediction without disclosing them! Check out our encrypted machine learning (EML) models applicable throughout chemical compound space. Hard work pulled off by @WeinreichJan and @ferchault arxiv.org/abs/2212.04322 #compchem #qml
Guido von Rudorff retweeted
The @PCCP Perspective in which 70 scientists discuss the status and future of DFT has been accepted for publication and posted on the journal pages: pubs.rsc.org/en/Content/Arti…. I am honored to be among the authors. Andy, Trygve and Andreas: Thanks for leading this effort! @MSUNatSci
Theoretical & computational chemistry and materials science communities may find this exchange of opinions on the current status and future of DFT interesting: doi.org/10.26434/chemrxiv-20…. Thank you Andy, Trygve, and Andreas for organizing the discussion and including my voice in it!
Guido von Rudorff retweeted
all pumped up about Simon's awesome contribution: Relative energies across #chemicalspace _without_ electronic contributions from alchemical integral transforms. He's literally warping real space in ways similar to special relativity theory! Amazing!
arxiv.org/abs/2203.13794
Guido von Rudorff retweeted
HOMO-LUMO gaps are not smooth in chemical space!! @mazouinb and Alexandre (now @corminboeuf_lab) just showed how to solve this problem using prior classification resulting in selected QML models HOMO-LUMO gaps with much improved data-efficiency! arxiv.org/abs/2110.02596
So proud!
Guido von Rudorff retweeted
'Data Enhanced Reaction Predictions in Chemical Space With Hammett's Equation' by Marco and @ferchault arxiv.org/abs/2004.14946: Augmentation with #machinelearning yields MP2 quality SN2 activation energies for only hundreds of training instances #COMPCHEM