Paul J. Atzberger

Professor, Mathematics and Mechanical Engineering, University of California Santa Barbara.

[Paul J. Atzberger @atzberg]

✴️ Pleased to share our latest work on Geometric Neural Operators (GNPs): Package & Papers ✴️

Point-Cloud Data, PDE Solvers, Pre-trained Foundation Models for Geometric Tasks

🔺Download ➡
https://t.co/kd5ghiLUJE

🔺More Information & Papers ➡
https://t.co/kd5ghiLUJE

[Paul J. Atzberger @atzberg]

✴️ Excited to share our latest research in PLOS Computational Biology!

"Protein Drift-Diffusion In Membranes With Non-Equilibrium Fluctuations" 🧬✨

https://t.co/V7oU1gQzVf

[Paul J. Atzberger @atzberg]

Thought I'd share our beginner-friendly JupyterNotebooks for learning the basics of tensors, broadcast rules, optimization, neural networks, and other topics. Used at UCSB for helping students get started on machine learning projects. GitHub link: https://t.co/19D6vbRa4n

[Paul J. Atzberger @atzberg]

A UCSB article discussing some of our work on geometric AI/ML and applications. https://t.co/EoL8uydJ2c Related Github: https://t.co/kd5ghiLUJE

[Paul J. Atzberger @atzberg]

Please to share our recent paper on Geometric Neural Operators (GNPs) for PDEs, shape identification, and other non-euclidean tasks. For more information, see @ResearchGate: https://t.co/kMW7XPiIjy

[Paul J. Atzberger @atzberg]

Pleased to share paper https://t.co/ytIAagtRnE on generative learning for non-linear dynamics and stochastic systems. Allows for purely sample-based learning, learning force-laws from 3D trajectory data, mechanical system parameters, and other tasks, https://t.co/IKZPn6LQkm.

[Paul J. Atzberger @atzberg]

Pleased to share our talk on Generative Machine Learning Approaches for Data-Driven Modeling of Dynamics, https://t.co/8G4RDBYNuN We discuss SDYN-GANs, adversarial learning, and related applications in scientific simulation. Additional information: https://t.co/IKZPn6LQkm

[Paul J. Atzberger @atzberg]

Pleased to share our fluids talk: Stochastic Immersed Boundary Methods, https://t.co/U5D7ciSSdl Covers theory & practice of simulation methods for mesoscopic fluid-structure interactions subject to thermal fluctuations for soft materials, complex fluids, and other applications.

[Paul J. Atzberger @atzberg]

Machine Learning Post-doc Position Available.

Projects at the interface of mathematical foundations, scientific computation, and related applications in sciences / engineering. Joint projects also with partners at Dept. of Energy. To apply go to https://t.co/Du8gV0N6ak

[Paul J. Atzberger @atzberg]

I am pleased to announce latest work with David Rower and Misha Padidar introducing surface fluctuating hydrodynamics methods for curved fluid interfaces. Preprint is available on arXiv at https://t.co/Q42aGv8LYr

[Paul J. Atzberger @atzberg]

I am pleased to announce latest work with Ben Gross, Nat Trask, and Paul Kuberry on meshfree solvers for hydrodynamic flows on curved surfaces. Preprint is available on arXiv at https://t.co/mJ3QGHoE0m

[Paul J. Atzberger @atzberg]

Developed and teaching two new #datascience courses this Fall at UCSB. An undergraduate course, Machine Learning: Foundations and Applications (MATH CS 120), https://t.co/u7IUf1Kd6x and a graduate course, Special Topics in Machine Learning (MATH 260J), https://t.co/YsedvVAm38

[Paul J. Atzberger @atzberg]

Thomas Fire from my drive on highway 101 yesterday ~4pm (Wed, Dec 6th) around Mussel Shoals. Returning to SB from NIPS conference in LA. #ThomasFire #NIPS2017