Marcell Stippinger

Journal club on PINN @ DCS, Wigner RCP

The AI ambassadors of HUN-REN Wigner RCP are organizing a journal club. Everyone interested in physics informed neural networks (PINN) from HUN-REN Wigner RCP or other HUN-REN or Hungarian academic institutions are welcome. To get an overview of the event series, click here.

Todays presentation

Presenter: Molnár Balázs (HUN-REN EK)

Title: PINNs for state estimation and control

Todays papers:

• State estimation and learning inverse mappings with
  1. PINN-Obs https://arxiv.org/abs/2507.06712,
  2. DeepONet https://arxiv.org/abs/1910.03193

In order to acheive:

Continue reading

Journal club on PINN @ DCS, Wigner RCP

The AI ambassadors of HUN-REN Wigner RCP are organizing a journal club. Everyone interested in physics informed neural networks (PINN) from HUN-REN Wigner RCP or other HUN-REN or Hungarian academic institutions are welcome. To get an overview of the event series, click here.

Todays presentation

Presenter: Kovács Péter (HUN-REN Wigner FK, KTO)

Title: Parciális differenciálegyenletek megoldása fotonikus hálókkal

How to join

It is preferred to read the selected paper in advance for a more in-depth discussion. For each paper we select a participant in advance who is asked to present the paper on the big srceen.

Continue reading

Journal club on PINN @ DCS, Wigner RCP

The AI ambassadors of HUN-REN Wigner RCP are organizing a journal club. Everyone interested in physics informed neural networks (PINN) from HUN-REN Wigner RCP or other HUN-REN or Hungarian academic institutions are welcome. To get an overview of the event series, click here.

Todays papers

7. Wang, Sifan, Shyam Sankaran, Hanwen Wang, and Paris Perdikaris. „An Expert’s Guide to Training Physics-Informed Neural Networks”.

• 2023, arXiv
• https://doi.org/10.48550/arXiv.2308.08468
• more theory, on boundary conditions

8. Krishnapriyan, Aditi S, Amir Gholami, Shandian Zhe, Robert M Kirby, and Michael W Mahoney. „Characterizing Possible Failure Modes in Physics-Informed Neural Networks”

• 2021, Advances in Neural Information Processing Systems 34: 26548-26560
• https://proceedings.neurips.cc/paper/2021/hash/df438e5206f31600e6ae4af72f2725f1-Abstract.html
• more advanced, regularization

Presented by: Ferenc Csikor

Continue reading

Journal club on PINN @ DCS, Wigner RCP

The AI ambassadors of HUN-REN Wigner RCP are organizing a journal club. Everyone interested in physics informed neural networks (PINN) from HUN-REN Wigner RCP or other HUN-REN or Hungarian academic institutions are welcome. To get an overview of the event series, click here.

Todays paper

We finish discussing

2. Raissi, M., P. Perdikaris and G.E. Karniadakis. „Physics-Informed Neural Networks: A Deep Learning Framework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations”.

• 2019, Journal of Computational Physics 378: 686–707.
• https://doi.org/10.1016/j.jcp.2018.10.045
• Code compatible with latest tensorflow: https://github.com/stippingerm/PINNs/

We briefly look into a paper that used generative AI for analytic/symbolic calculations

Continue reading

Journal club on PINN @ DCS, Wigner RCP

The AI ambassadors of HUN-REN Wigner RCP are organizing a journal club. Everyone interested in physics informed neural networks (PINN) from HUN-REN Wigner RCP or other HUN-REN or Hungarian academic institutions are welcome. To get an overview of the event series, click here.

Todays paper

2. Raissi, M., P. Perdikaris, and G.E. Karniadakis. „Physics-Informed Neural Networks: A Deep Learning Framework for Solving Forward and Inverse Problems Involving Nonlinear Partial Differential Equations”.

• 2019, Journal of Computational Physics 378: 686–707.
• https://doi.org/10.1016/j.jcp.2018.10.045
• Code compatible with latest tensorflow: https://github.com/stippingerm/PINNs/

Presented by: Marcell Stippinger

Continue reading

Journal club on PINN @ DCS, Wigner RCP

The AI ambassadors of HUN-REN Wigner RCP are organizing a journal club. Everyone interested in physics informed neural networks (PINN) from HUN-REN Wigner RCP or other HUN-REN or Hungarian academic institutions are welcome. To get an overview of the event series, click here.

Todays paper

Cuomo, Salvatore, Vincenzo Schiano Di Cola, Fabio Giampaolo, Gianluigi Rozza, Maziar Raissi, és Francesco Piccialli. „Scientific Machine Learning Through Physics–Informed Neural Networks: Where We Are and What’s Next”.

Continue reading

Ai4Science program intro

This seminar gives you a unique opportunity to both present your work involving AI and ML methods, and meet researchers with similar interest. We aim to bring closer the researchers to the members of the AI4Science technical support team.

Registration: indico page

Time: 14:00-16:30, Friday, December 6, 2024

Location: Department of Computational Sciences, HUN-REN Wigner RCP, Building 6, Floor 2

Intended audience: members of the HUN-REN Wigner RCP

Programme (hun) Szabó Gergő (BME) nálunk tart előadást az AI4Science programban végzett mesterséges intelligencia támogatásról. Beszél arról, hogy a HUN-REN kutatóhálózatban milyen kutatási projektek megvalósításában segít mesterséges intelligencia szakértőként, illetve, hogy mi a támogatás igénylésének folyamata.

Continue reading

Large language models in programming @ DCS, Wigner RCP

This event is an updated version of our previous presentation on effectively using AI tools for coding, understanding code, and writing documentation. If you missed the previous talk or would like a refresher, please join us.

This session partly builds on the generative AI course offered by HUN-REN to its researchers, with adaptations tailored specifically for the physics research community.

Registration: indico page

Time: 14:00-15:30, Friday, November 15, 2024

Location: Department of Computational Sciences, HUN-REN Wigner RCP, Building 6, Floor 2

Continue reading

Large language models in programming @ DCS, Wigner RCP

This event, first of a series, kicks off the HUN-REN Wigner seminars on AI and ML, best practices based on the HUN-REN AI 4 Science Programme.

If you are interested in the efficient use of AI tools for coding understanding code and writing documentation, please come along. This session partly builds on the generative AI course that HUN-REN offers to its researcher community (i.e., to you).

Registration: indico page

Time: 15:30-17:00, Monday, October 21, 2024

Continue reading

The Nobel Prize in Chemistry 2024

The Nobel Prize in Chemistry 2024 was divided, one half awarded to David Baker “for computational protein design”, the other half jointly to Demis Hassabis and John M. Jumper “for protein structure prediction”.

Continue reading