Mario 🇺🇦 💙💛 🇮🇱 + Scientific Machine Learning
@ScientificML
Followers
413
Following
1K
Media
37
Statuses
2K
{ Visual & Tensor Computing; Grad student in ML4Science; Vision, Deep Neural Networks, Optimization, and SciML Simulation; Python/NumPy, PyTorch, JAX, *nix,git}
Mars
Joined February 2010
RT @AnimaAnandkumar: In a recent interview I talk about what it takes for AI to make new scientific discoveries. tldr: it won’t be just LLM….
newindiaabroad.com
Learn how Indian American professor Anima Anandkumar is revolutionizing the world of artificial intelligence to drive new scientific discoveries. Explore her cutting-edge research and innovative...
0
21
0
More details can be found on the project web page:
0
0
0
Please join this timely workshop on Multi-scale Modeling for (Physical & Chemical) Sciences and Engineering problems. We propose LOGLO-FNO, an improvement to the popular FNO arch, specifically targeted towards modeling high frequencies in turbulence simulations. #ai4science #pde.
🚀Continuing the spotlight series with the next .@iclr_conf MLMP 2025 Oral presentation!.📝LOGLO-FNO: Efficient Learning of Local and Global Features in Fourier Neural Operators.📷 Join us on April 27 at #ICLR2025!.#AI #ML #ICLR #AI4Science.
1
2
4
RT @DHolzmueller: Poster: Hall 3 + Hall 2B #32, 10am Singapore time.Paper:
arxiv.org
Solving partial differential equations (PDEs) is a fundamental problem in science and engineering. While neural PDE solvers can be more efficient than established numerical solvers, they often...
0
1
0
RT @DHolzmueller: 🚨ICLR poster in 1.5 hours, presented by Daniel Musekamp:.Can active learning help to generate better datasets for neural….
0
7
0
RT @artemmoskalev: Thrilled to receive the outstanding paper award together with @Mangal_Prakash_ for our work on SE(3)-Hyena https://t.co/….
0
6
0
RT @bernat_font: The WaterLily.jl preprint is now available on arXiv! 📰.WaterLily is a #CFD solver written in the #….
0
55
0
The proposed VCNeF architecture marries the two worlds, #NeuralOperators & #NeuralFields, and outperforms several SOTA methods such as Transformers, Neural Fields, Graph Neural Networks, and Fourier Neural Operators for solving parametric PDEs, e.g. Compressible Navier-Stokes.
0
0
0
To develop a model that encompasses these ideal characteristics, we propose VCNeF. This linear transformer-based conditional neural field continuously solves PDEs in space and time, endowing the model with spatial and temporal zero-shot super-resolution capabilities.
1
0
0
Neural Nets are increasingly used to solve PDEs. Despite recent advances, current methods lack several characteristics of an ideal neural PDE solver, viz. generalization to different initial conditions, PDE parameter values, and spatial and temporal super-resolution capabilities.
1
0
1
📢#AI4Science Vectorized Conditional Neural Field (VCNeF) for Time-dependent Parameterized Partial Differential Equations at #ICML2024. VCNeF marries Neural Operators & Neural Fields. Project: Code: Paper:
7
7
52
RT @DimaZeniuk: “Starship is the kind of thing that makes people from little kids to seniors excited about the future.”. — Elon Musk https:….
0
3K
0
RT @maksym_andr: Super excited to share that I successfully defended my PhD thesis "Understanding Generalization and Robustness in Modern D….
0
12
0
RT @AI4scienceTalks: 📢#AI4Science Talk on 12.02.24 at 15:00 (CET) / 09:00 (ET) on “Stochastic Optimal Control for Collective Variable Free….
0
5
0
RT @BorisHanin: 🚨Princeton ML Theory Summer School🚨. Aug 6 - 15, 2024. Speakers:. * F. Krzakala/L. Zdeborova @KrzakalaF @zdeborov….
0
24
0
RT @AI4scienceTalks: Please join us tomorrow for this interesting presentation if you're interested to learn about self-supervised learning….
0
2
0
RT @SebastienBubeck: My group is hiring a large cohort of interns for the summer of 2024 to work on the Foundations of Large Language Model….
0
48
0
RT @AI4scienceTalks: 📢#AI4Science Talk on 30.11.23 at 15:00 (CET) / 09:00 (ET) on “Self-Supervised Learning with Lie Symmetries for Partial….
0
4
0