🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

You can stay tuned on our work in #AI, #CFD, #FluidMechanics and #Sustainability in our new #GroupWeb!! We will post updates on new #publications, #positions and possible #collaborations!! #VinuesaLab vinuesalab.com

Yesterday we had a #seminar in the AC& #machinelearning thematic area at KTH Royal Institute of Technology. We were delighted to have Steven Brunton with us, who gave a very exciting presentation and discussed the potential of #deeplearningmodels and #ai in #fluidmechanics!! play.kth.se/media/AC%26ML+…

Can we solve RANS equations without modeling assumptions? Yes, we can! But we need some DATA. Check out our paper on solving RANS equations using Physics-Informed Neural Networks (#PINNs). With Ricardo Vinuesa and Philipp Schlatter. arxiv.org/abs/2107.10711

#Research #Seminar with Hamidreza Eivazi for the #CRUNCH group of Prof. George Karniadakis! #deeplearning, #turbulence and #PhysicsInformed #neuralnetworks (#PINNs) for #RANS #simulations. #Join via #Zoom! August 27, 12pm ET/18 CEST Zoom link: brown.zoom.us/j/97005015812 Steven Brunton

Very interesting paper by Hamidreza Eivazi, Soledad Le Clainche, Sergio Hoyas & Ricardo Vinuesa! arxiv.org/abs/2109.01514 POD decomposes a flow in orthogonal modes (by definition!), but it requires much more modes to reconstruct the flow to good accuracy wrt nonlinear 1/n

In our latest #article, Steven Brunton and I we #review the areas within #CompuationalFluidDynamics (#CFD) where #ML/#AI can help in terms of #Acceleration and/or #Improved #Physical #Knowledge. #video by Steven Brunton: youtube.com/watch?v=IXMSOS… #Preprint: arxiv.org/abs/2110.02085

The #NeurIPS2021 workshop on Deep-Learning-meets-Differential-Equations is coming up! (14th)⚡️ For the panel discussion (on solving DEs with DL), submit questions in advance: forms.gle/j37v5zmck1k1ps… Note the schedule: accessible from around the world, not just US timezone!⏰️

In order to better #communicate our #research, we will share a series of #videos, starting with one on #ML (#MachineLearning), #CFD (#computationalfluiddynamics) and #Autoencoders for #ModalDecomposition! #Link: youtube.com/channel/UCx_UR… #AI #Turbulence #FluidMechanics #DataDriven

#JAX, #JIT, #AD (automatic differentiation), #Fortran-#Python communication, and more for fast multiscale finite-element computations via deep neural networks. Check out our new article: doi.org/10.3390/mca280… #Code and #data: github.com/HamidrezaEiv/F…

#Thunderstorm #Prediction is a critical task within #AirTraffic #Management (#ATM). In our latest #paper, we use #DeepLearning to enhance existing methods to predict thunderstorms, thus enabling much more efficient Pre-Tactical #ATM!! #Article: sciencedirect.com/science/articl…

⚡ Can we predict #thunderstorms? ⛈️ Taking the opportunity of today's International Civil Aviation Day, we share an interesting scientific paper crafted by our partners from KTH Royal Institute of Technology about how #DeepLearning can help to achieve better #AirTraffic Management! ICAO

Nonlinear model reduction for operator learning. arxiv.org/abs/2403.18735

Excited to share my latest research on “Nonlinear model reduction for operator learning” at #ICLR2024 next week. Poster Session: Wed 8 May 10:45 a.m. Halle B #320 Oral Session: Fri 10 May 1:15 p.m. Halle A 8 - 9 ➤ Paper: openreview.net/forum?id=Jw6TU… ➤ Code: github.com/HamidrezaEiv/K…

Had a fantastic time at ICLR 2026! Loved chatting AI for science with brilliant minds I've followed for years and making new friends. Inspiring experience! #ICLR2024 #AI #Science #Physics

Can one use #ML to help correct #Experimental #Measurements in #Fluid #Mechanics? In our work published in IOP Publishing #Meausurement #Science and #Technology, we use #PINNs for #SuperResolution, to reduce #Noise and to correct #HotWire measurements! iopscience.iop.org/article/10.108…

Our paper, "DiffBatt: A #DiffusionModel for Battery Degradation Prediction and Synthesis," has been accepted for the Foundation Models for Science Workshop at #NeurIPS 2024! Paper: arxiv.org/abs/2410.23893 Codes, pre-trained models and data: github.com/HamidrezaEiv/D…

I will present our work on foundation modeling for #battery degradation at the #NeurIPS2024 Foundation Models for Science workshop. Come check out our poster on Sunday, December 15th, from 3:00 PM to 3:30 PM PST. Poster: lnkd.in/efCPPGU5 Paper: lnkd.in/eTSCi55b