tum-pbs/pbdl-book
An interactive Jupyter book on physics-based deep learning covering neural networks for scientific simulations, differentiable physics, and probabilistic generative models.
Not currently ranked — collecting fresh signals.
star history
This repository hosts an interactive book on physics-based deep learning, emphasizing practical applications over pure theory. Every concept is paired with executable Jupyter notebooks demonstrating deep learning for physical simulations. The content covers traditional supervised learning, physics-informed loss constraints, differentiable simulations, diffusion-based probabilistic methods for generative AI, reinforcement learning with simulators, and advanced neural network architectures for scientific computing.
Frequently asked
- What is tum-pbs/pbdl-book?
- An interactive Jupyter book on physics-based deep learning covering neural networks for scientific simulations, differentiable physics, and probabilistic generative models.
- Is pbdl-book open source?
- Yes — tum-pbs/pbdl-book is an open-source project tracked on heatdrop.
- What language is pbdl-book written in?
- tum-pbs/pbdl-book is primarily written in Jupyter Notebook.
- How popular is pbdl-book?
- tum-pbs/pbdl-book has 1.4k stars on GitHub.
- Where can I find pbdl-book?
- tum-pbs/pbdl-book is on GitHub at https://github.com/tum-pbs/pbdl-book.