AI4Science Laboratory
@ai4science_lab
Laboratory for Artificial Intelligence for Scientific Discovery at @UvA_Amsterdam. #MachineLearning #AI
قد يعجبك
Large Language Models & Flow Matching & Materials? Bound to be interesting! Very impressive results too! (50% more stable unique and novel materials than previous works!) A nice continuation of works together with @AmlabUva , @ai4science_lab, @AIatMeta, @OpenCatalyst.
I’m excited to share our latest work on generative models for materials called FlowLLM. FlowLLM combines Large Language Models and Riemannian Flow Matching in a simple, yet surprisingly effective way for generating materials. arxiv.org/abs/2410.23405 @bkmi13 @RickyTQChen @bwood_m
🚀 Interested in time series generation?⏲️Excited to share my @GoogleDeepMind Amsterdam student researcher project: Rolling Diffusion Models! arxiv.org/abs/2402.09470 (to appear at ICML 2024) Thanks for the great collaboration @emiel_hoogeboom, @JonathanHeek, @TimSalimans! 🧵1/4
New exciting work by the AI4Science lab: geometric convolutional neural networks that work in any dimension and even pseudo-Euclidean geometries!
Excited to introduce Clifford-Steerable CNNs: a framework that expands equivariant CNNs to pseudo-Euclidean groups, including the Poincaré group - the group of isometries of spacetime! Joint work w/ @djjruhe, @maurice_weiler, @__alucic, @jo_brandstetter, and Patrick Forré 1/12
Congrats @djjruhe! Really great work!
Our paper has been accepted as an oral to @NeurIPSConf ! 😃🥳 📜Updated paper at arxiv.org/abs/2305.11141 💻Updated code (including Colab tutorial) at github.com/DavidRuhe/clif…
Hey materials and molecules people! Check out this really cool API for the @OpenCatalyst In just a few lines you can interact your adsorbate with a surface!
Today, we're releasing ocpapi, a python client for accessing the API powering the Open Catalyst Demo.
Nice opportunity to show off one of your methods!
New work on Clifford algebra equivariant networks by @djjruhe, @jo_brandstetter and Patrick Forré
Together with the amazing @jo_brandstetter and Patrick Forré, we present Clifford Group Equivariant Neural Networks: a new E(n) steerable equivariant neural architecture that respects rotations, reflections, and more symmetries that operates on multivectors!
Yesterday, I presented our work "Simulation-based Inference with the Generalized Kullback-Leibler Divergence" which combines ratio and posterior estimation methods! It can even do it within the same (hybrid) model. openreview.net/forum?id=tPb1F…
First official release of the DeepQMC code for Quantum Monte Carlo with #NeuralNetworks wave functions, including architectures like PauliNet, FermiNet and DeepErwin. By Zeno Schätzle, @PBerntSzab1, Matej Mezera and @jhrmnn arxiv.org/abs/2307.14123
We are excited to announce that the Machine Learning For The Physical Sciences workshop will be back at @NeurIPSConf for its 7th edition! Updates will be announced through this Twitter account. Follow us ✨ Our website: ml4physicalsciences.github.io/2023/ #ML4PhysicalSciences
Are you at @icmlconf and interested in chatting about Clifford geometric algebras? Come by at our poster #534 from 11am onward!
Great work @bkmi13!
Yesterday, our method of Contrastive Neural Ratio Estimation was pulled into the main branch of lampe github.com/francois-rozet…! lampe is a very promising #sbi library by @FrancoisRozet from @glouppe's group in Liège. It takes a bare-bones approach, just like pytorch. Take a look!
Lots of transition path sampling in one tweet at the Dutch Soft Matter Day in Amsterdam
Very proud of our very own Jim Boelrijk's work on bayesian optimization! Great work!
Delighted to be in Kigali for #ICLR2023! If you happened to have missed my poster, yet still want to link up on scaling multi-objective optimization to more objectives, feel free to hit me up here or on Whova. Happy to chat over coffee and pastries😋 openreview.net/forum?id=fSa5I…
It was rewarding to collaborate with @ArnaudDelaunoy and @glouppe on this topic! I'm proud of our work together! We aim to extend the toolbox for estimating conservative posteriors in simulation-based inference with the balance criterion. #sbi
Balancing has empirically proven to produce more conservative posterior approximations. However, this technique was limited to ratio-based algorithms. This work extends its applicability to all methods that provide posterior density estimation. arxiv.org/abs/2304.10978
Very proud of @gustin_julien and Norman's latest work on simulation-based inference for robotic grasping! 🤖🤏☕️ #sbi
🚨Looking for 2 postdocs to join @UvA_Science @ai4science_lab 4 sustainable molecules & materials! Salt hydrates 4 thermal storage (with Shahidzadeh, Woutersen) Plant proteins 4 sustainable food (Bolhuis, van Hoof, Jabbari-Farouji, Quattrocchio, Schall) vacatures.uva.nl/UvA/job/4-Post…
United States الاتجاهات
- 1. No Kings 470K posts
- 2. Ange 57.4K posts
- 3. Dork Cult Protest Day 5,548 posts
- 4. Chelsea 143K posts
- 5. Good Saturday 33.9K posts
- 6. #Talus_Labs N/A
- 7. Nuno 8,481 posts
- 8. Forest 127K posts
- 9. #Caturday 3,734 posts
- 10. #SaturdayVibes 4,494 posts
- 11. Gameday 12.8K posts
- 12. #NFOCHE 32.9K posts
- 13. Marinakis 6,292 posts
- 14. Emiru 15.4K posts
- 15. Reece James 16.5K posts
- 16. Neto 32.2K posts
- 17. Massie 44.3K posts
- 18. Garnacho 19.9K posts
- 19. Ohtani 254K posts
- 20. Malo Gusto 3,856 posts
قد يعجبك
-
Frank Noe
@FrankNoeBerlin -
Johannes Brandstetter
@jo_brandstetter -
Maurice Weiler
@maurice_weiler -
David Ruhe
@djjruhe -
Gianni De Fabritiis
@gdefabritiis -
Efstratios Gavves
@egavves -
Sindy Löwe
@sindy_loewe -
Benjamin Kurt Miller
@bkmi13 -
Phillip Lippe
@phillip_lippe -
Giuseppe Cavaliere
@CavaliereGiu -
AI for Science
@AI_for_Science -
Simon Batzner
@simonbatzner -
Sharvaree Vadgama
@SharvVadgama -
Erik Bekkers
@erikjbekkers -
Deepak Pathak
@pathak2206
Something went wrong.
Something went wrong.