General relativity 🤝 neural fields
This simulation of a black hole is coming from our neural networks 🚀
We introduce Einstein Fields, a compact NN representation for 4D numerical relativity. EinFields are designed to handle the tensorial properties of GR and its derivatives.
Today marks a big milestone for us at Emmi AI. We’ve raised a €15M seed round, backed by 3VC, Speedinvest, Serena, and PUSH. Let’s build the future of Physics AI together!
Super hyped to share NeuralDEM -- the first real-time simulation of industrial particulate flows. NeuralDEM replaces Discrete Element Method (DEM) routines and coupled (CFD-DEM) multiphysics simulations. 🧵
📜: https://t.co/JH4PDpth5g
🖥️: https://t.co/VEsawzd9IV
Interesting in scaling up neural operators? Happy to announce that Universal Physics Transformers (UPT) -- a scalable framework for neural operators is accepted at #neurips2024.
Paper: https://t.co/fMAVb42UjD
Project page: https://t.co/8Vj2scLeeC
We introduce Geometry-Informed Neural Networks to train shape generative models
without any data (!!), combining learning under constraints, neural fields as a suitable representation, and generating diverse solutions to under-determined problems:
🖥️: https://t.co/qRbJ9SXuc0
New work on Geometric Clifford Algebra Networks (GCANs). We propose geometric templates for modeling dynamical systems. A 🧵on geometric / Clifford algebras, and symmetry group transformations in neural networks.
📜https://t.co/ugW5HAgFi6
Our paper "Addressing Parameter Choice Issues in Unsupervised Domain Adaptation by Aggregation" has been selected for oral presentation (notable-top-5%) at #ICLR2023 https://t.co/M0SnPF6Ufq. [1/n]
New work on how to construct neural network layers on composite objects of scalars, vectors, bivectors, … --> multivectors! Via Clifford algebras, we generalize convolution and Fourier transforms to multivectors, especially relevant for PDE modeling:
https://t.co/51F2T3gnX0
Interested in few-shot learning beyond miniImageNet? Deep nets often struggle to predict systems with varying parameters when data are scarce. Our new few-shot learning method SubGD is here to help!
https://t.co/kamNVNkwtH 🧵 1/5
Are you interested in graph structured data?
Do you want to include geometry and physics to boost your GNNs?
Check out our paper on Steerable E(3) Equivariant Graph Neural Networks.
Joint work with @robdhess @ElisevanderPol@erikjbekkers@wellingmax
https://t.co/exBcDhQ44k
Great work lead by @AndreasMayr11. Granular flows learn to move inside complex geometric objects without any handcrafted restrictions. Paper: https://t.co/qKIBDKZVJZ Blog post: https://t.co/eWL1A7Awuv
Fantastic work by the AMLAB team Johannes Brandstetter @jbrandi6, Rob Hesselink @robdhess, Elise van der Pol @ElisevanderPol and Erik Bekkers @erikjbekkers. We trained a fully steerable equivariant GNN to reach the top of the leaderboard. Congrats team.
Our paper "Hopfield Networks is All You Need" is accepted at #ICLR2021. Time to give some talks :) I am very honored to present our research today at the great platform of @ml_collective@savvyRL (https://t.co/vb2n5cMjcL).
Want to hear what @DimaKrotov and Sepp Hochreiter say about Modern Hopfield Networks? Then join tomorrow's discussion @NeurIPSConf hosted by @IARAInews
Tackling an extremely massive multiple instance learning problem in immunology with deep learning and attention (modern Hopfield networks!)
Check out our spotlight and poster at #neurips2020! 🙂
Dec 09, 07:10-07:20 (Spotlight), 09:00-11:00 (Poster) AM PST
https://t.co/PwQrdCgqkG