CARLA 2024

Invited Speaker

John Doe Designer
John Doe Designer

Day

Monday September 18, 2023

Time:

14:00 to 17:30 hours

Instructors:

Pedro Mario Cruz e Silva, NVIDIA Corporation

Room

4

ROOM 4. Introduction to Physics-Informed Machine Learning with Modulus

High-fidelity simulations in science and engineering are computationally expensive and time-prohibitive for quick iterative use cases, from design analysis to optimization. NVIDIA Modulus, the physics machine learning platform, turbocharges such use cases by building physics-based deep learning models that are 100,000x faster than traditional methods and offer high-fidelity simulation results.

Upon completion, you will have an understanding of the various building blocks of Modulus and the basics of physics-informed deep learning. You’ll also have an understanding of how the modulus framework integrates with the overall Omniverse Platform.

Student's prerequisites

Familiarity with the Python Programming Language

    •  An understanding of partial differential equations and their use in physics
    •  Familiarity with machine learning concepts like training and inference

Program:

  • Use the Modulus API
  • Solve data-driven and physics-driven problems using Modulus
  • Utilize techniques that Modulus offers to solve problems ranging from deep learning to modeling multi-physics simulations systems