Miguel Liu-Schiaffini

🚀 Thrilled to launch DeepScholar, an openly-accessible DeepResearch system we've been building at Berkeley & Stanford. DeepScholar efficiently processes 100s of articles, demonstrating strong long-form research synthesis capabilities, competitive with OpenAI's DR, while running up to 2x faster! Try it out: deep-scholar.vercel.app

Neural Operators – Deep learning at any resolution Extending neural networks to function spaces: While many phenomena are inherently described by functions, neural networks define vector-to-vector mappings that rely on fixed discretization of the input and output. Neural Operators instead define learnable function-to-function mappings that guarantee consistent predictions across different discretization of the input and output functions. By respecting the functional nature of the data, neural operators can achieve improved performance and generalization. Translating the success of deep learning to operator learning: Careful engineering of neural architectures has been a key factor in deep learning’s success. Translating these architectures to neural operators is crucial for operator learning to enjoy the same empirical optimizations. Key principles for constructing Neural Operators: *Recipes for converting popular architectures (CNNs, GNNs, transformers, etc.) into Neural Operators *Guidance for practitioners arxiv.org/abs/2506.10973 github.com/neuraloperator @julberner @mliuschi @JeanKossaifi Valentin Duruisseaux Boris Bonev @Azizzadenesheli @caltech

2024 was a pivotal year for AI+Science. Our team made exciting contributions. Here are some highlights: tensorlab.cms.caltech.edu/users/anima/20… 1. Neural Operators as a unifying AI framework for modeling multi-scale processes. We got to write a perspective article in @NatRevPhys and released an open-source library that has been widely adopted. 2. FourCastNet, our Neural-Operator-based model, featured in the @WHOSTP PCAST report on AI+Science. Its speed enables large ensembles to obtain unprecedented predictions of extreme weather events. 3. Designed a novel medical catheter using Neural Operator-based inverse design. 4. Used Neural Operators with reinforcement learning for turbulence stabilization and wall friction reduction. 5. Many other applications where Neural Operators have shown a big impact: nuclear fusion, computational imaging, Carbon capture and storage etc. 6. Gave a @TEDTalks highlighting AI+Science. 7. Exascale training of genome-scale language models for protein design that was recognized as a finalist for the ACM Gordon-Bell prize. 8. State-of-the-art protein-ligand structure prediction that can handle changing protein conformations that Alphafold and other methods cannot. 9. AI+Math innovations with LLM + Lean for theorem proving. 10. Hardware-efficient large-scale training such as gradient projections and mini-sequence transformers. 11. Many honors such as @iitmadras Distinguished Alumnus award, and @BlavatnikAwards

Introducing NeuralOperator 1.0: a Python library that aims at democratizing neural operators for scientific applications by providing all the tools for learning neural operators in PyTorch : state-of-the-art models, built-in trainers for quick starting and modular neural operator blocks for advanced used in your own workflow or to build new architectures.

This release was long in the making and the result of a large group effort. Check out our white paper: arxiv.org/abs/2412.10354 With @ZongyiLiCaltech, Nikola Kovachki, David Pitt, @mliuschi, @Robertljg, Boris Bonev, @Azizzadenesheli, @julberner and @AnimaAnandkumar

Excited to present at NeurIPS 2024! 🎉 We propose a unified neural operator for Compressed Sensing MRI, adapting to multiple undersampling patterns/rates, with 11% SSIM & 4dB PSNR gains. Full paper & code: armeet.ca/nomri #NeurIPS2024 #ML #MRI #NeuralOperators

#NeurIPS I am on the 2024-25 job market seeking faculty positions and postdocs! My goal is to advance AI for scientific computing and discovery. I develop neural operators for partial differential equations (PDEs) with applications in fluid, solid, and earth science.

📢 Come by our #ICML2024 poster "Neural Operators with Localized Integral and Differential Kernels"! 🚀 We propose two extensions of standard convolutional layers to operator learning! Location: Poster #216, 11:30 - 13:00 CEST @AnimaAnandkumar @julberner @Azizzadenesheli

Drop by our #ICML2024 posters to chat about neural operators and PDE solvers 1⃣Solving Poisson Eqs. using Neural Walk-on-Spheres 2⃣DPOT: Auto-Regressive Denoising Operator Transformer for Large-Scale PDE Pre-Training 3⃣Neural Operators w/ Localized Integral & Differential Kernels

📢Check out our work at #ICML2024 on solving high-dimensional Poisson equations using neural Walk-on-Spheres (WoS)  ⚡️ faster and more accurate than PINNs, Deep Ritz, and other SDE-based methods Session: Tue 23 Jul 11:30 am - 1 pm, Hall C 4-9 #2811 @julberner @AnimaAnandkumar

@AnimaAnandkumar @Caltech Thanks!! 😊

Congratulations to @caltech undergrads on winning the Goldwater scholarship including Miguel who has been making key contributions to neural operators caltech.edu/about/news/gol…