Free Resource on AI for Physical Simulations

September 27, 2021

The academics at the Thuerey Group have made a useful book on artificial intelligence operations and smart software applications available online. The Physics-Based Deep Learning Book is a comprehensive yet practical introduction to machine learning for physical simulations. Included are code examples presented via Jupyter notebooks. The book’s introduction includes this passage:

“People who are unfamiliar with DL methods often associate neural networks with black boxes, and see the training processes as something that is beyond the grasp of human understanding. However, these viewpoints typically stem from relying on hearsay and not dealing with the topic enough. Rather, the situation is a very common one in science: we are facing a new class of methods, and ‘all the gritty details’ are not yet fully worked out. However, this is pretty common for scientific advances. … Thus, it is important to be aware of the fact that – in a way – there is nothing magical or otherworldly to deep learning methods. They’re simply another set of numerical tools. That being said, they’re clearly fairly new, and right now definitely the most powerful set of tools we have for non-linear problems. Just because all the details aren’t fully worked out and nicely written up, that shouldn’t stop us from including these powerful methods in our numerical toolbox.”

This virtual tome would be a good place to start doing just that. Interested readers may want to begin studying it right away or bookmark it for later. Also see the Thuerey Group’s other publications for more information on numerical methods for deep-learning physics simulations.

Cynthia Murrell, September 27, 2021


Comments are closed.

  • Archives

  • Recent Posts

  • Meta