Modeling Methods III: Deep Learning and Physics-Informed Learning

Presentations

• The effect of physical constraints on the loss function landscapes of deep learning models

  ORAL

  Nov. 20, 2023, 8:00 AM – Nov. 20, 2023, 8:13 AM

  Presenters

  • Manuel Cabral

    • TU Delft

  Authors

  • Manuel Cabral

    • TU Delft

  • Bernat Font

    • Barcelona Supercomputing Center
    • Barcelona Super Computing Center - Centro Nacional de Supercomputación (BSC-CNS), Spain

  • Gabriel D Weymouth

    • TU Delft

  View abstract →

• On application of Physics-Informed Neural Networks to Improve Noisy Data of Incompressible Flows

  ORAL

  Nov. 20, 2023, 8:13 AM – Nov. 20, 2023, 8:26 AM

  Presenters

  • Abdelrahman A Elmaradny

    • University of California Irvine

  Authors

  • Abdelrahman A Elmaradny

    • University of California Irvine

  • Ahmed Atallah

    • University of California, San Diego

  • Yasaman Farsiani

    • University of California Irvine

  • Haithem E Taha

    • UC Irvine
    • University of California Irvine

  • Arash Kheradvar

    • University of California Irvine

  View abstract →

• Explainable deep learning for fluid dynamics using a Fourier-wavelet analysis framework

  ORAL

  Nov. 20, 2023, 8:26 AM – Nov. 20, 2023, 8:39 AM

  Publication: Subel, Guan, Chattopadhyay and Hassnazadeh, Explaining the physics of transfer learning in data-driven turbulence modeling, PNAS Nexus, Volume 2, Issue 3, March 2023

  Presenters

  • Pedram Hassanzadeh

    • Rice University
    • University of Chicago

  Authors

  • Pedram Hassanzadeh

    • Rice University
    • University of Chicago

  • Ashesh K Chattopadhyay

    • University of California, Santa Cruz

  • Yifei Guan

    • Rice University

  • Hamid Pahlavan

    • Rice U

  • Adam Subel

    • New York University

  View abstract →

• Reduced-order modeling of fluid flows with transformers

  ORAL

  Nov. 20, 2023, 8:39 AM – Nov. 20, 2023, 8:52 AM

  Publication: https://doi.org/10.1063/5.0151515

  Presenters

  • AmirPouya Hemmasian

    • Carnegie Mellon University

  Authors

  • AmirPouya Hemmasian

    • Carnegie Mellon University

  • Amir Barati Farimani

    • Carnegie Mellon University

  View abstract →

• Development of reduced order modeling-based linear system extracting method for efficient data handling with a minimal nonlinearity

  ORAL

  Nov. 20, 2023, 8:52 AM – Nov. 20, 2023, 9:05 AM

  Publication: Planned to submit to arXiv.

  Presenters

  • Takeru Ishize

    • Keio university

  Authors

  • Takeru Ishize

    • Keio university

  • Koji Fukagata

    • Keio University
    • Keio Univ

  View abstract →

• Residual-based physics-informed transfer learning (RePIT) strategy to accelerate unsteady fluid flow simulations

  ORAL

  Nov. 20, 2023, 9:05 AM – Nov. 20, 2023, 9:18 AM

  Presenters

  • Joongoo Jeon

    • Jeonbuk National University

  Authors

  • Joongoo Jeon

    • Jeonbuk National University

  • Juhyeong Lee

    • Hanyang University

  • Ricardo Vinuesa

    • KTH (Royal Institute of Technology)
    • KTH Royal Institute of Technology

  • Sung Joong Kim

    • Hanyang University

  View abstract →

• Improving Neural Operators with Physics Informed Token Transformers

  ORAL

  Nov. 20, 2023, 9:18 AM – Nov. 20, 2023, 9:31 AM

  Publication: "Physics Informed Token Transformer" available on ArXiv: https://arxiv.org/abs/2305.08757 and in submission at APL Machine Learning

  Presenters

  • Cooper Lorsung

    • Carnegie Mellon University

  Authors

  • Cooper Lorsung

    • Carnegie Mellon University

  • Zijie Li

    • Carnegie Mellon University

  • Amir Barati Farimani

    • Carnegie Mellon University

  View abstract →

• Dimensional compression and reconstruction for unstructured finite volume meshes via geometric deep learning

  ORAL

  Nov. 20, 2023, 9:31 AM – Nov. 20, 2023, 9:44 AM

  Presenters

  • Liam K Magargal

    • Lehigh University

  Authors

  • Liam K Magargal

    • Lehigh University

  • Parisa Khodabakhshi

    • Lehigh University

  • Steven N Rodriguez

    • United States Naval Research Laboratory

  • Justin W Jaworski

    • Virginia Tech

  • John Michopoulos

    • United States Naval Research Laboratory

  View abstract →

• Physics-informed neural network for enhancement of weather forecasts

  ORAL

  Nov. 20, 2023, 9:44 AM – Nov. 20, 2023, 9:57 AM

  Publication: Planned article in writing process under same title

  Presenters

  • Alvaro Moreno Soto

    • Universidad Carlos III de Madrid

  Authors

  • Alvaro Moreno Soto

    • Universidad Carlos III de Madrid

  • Alejandro Cervantes

    • Universidad Carlos III de Madrid / Universidad Internacional de La Rioja

  • Manuel Soler

    • Universidad Carlos III de Madrid

  View abstract →

• Using self-adaptive physics-informed learning to estimate orographic gravity waves

  ORAL

  Nov. 20, 2023, 9:57 AM – Nov. 20, 2023, 10:10 AM

  Presenters

  • Thi Nguyen Khoa Nguyen

    • ENS Paris-Saclay

  Authors

  • Thi Nguyen Khoa Nguyen

    • ENS Paris-Saclay

  • Christophe Millet

    • CEA, DAM, DIF, F-91297 Arpajon, France

  • Thibault Dairay

    • Michelin

  • Raphaël Meunier

    • Michelin

  • Mathilde Mougeot

    • ENSIIE / ENS Paris-Saclay

  View abstract →

• Machine-learned reduced order modeling toward an effective flow control framework

  ORAL

  Nov. 20, 2023, 10:10 AM – Nov. 20, 2023, 10:23 AM

  Publication: Planned to submit to arXiv

  Presenters

  • Hiroshi Omichi

    • Keio University

  Authors

  • Hiroshi Omichi

    • Keio University

  • Takeru Ishize

    • Keio university

  • Koji Fukagata

    • Keio University
    • Keio Univ

  View abstract →

• A Data-Free Partial Differential Equation Solver Based on Physics-Informed Neural Networks (PINN): FDM-PINN

  ORAL

  Nov. 20, 2023, 10:23 AM – Nov. 20, 2023, 10:36 AM

  Presenters

  • Xiaoyu Tang

    • Northeastern University
    • Northeastern university

  Authors

  • Xiaoyu Tang

    • Northeastern University
    • Northeastern university

  • Boqian Yan

    • Northeastern.edu

  View abstract →