Physics-Based Deep Learning (Online Book)
This open book by Nils Thuerey, Philipp Holl and colleagues organizes the field into a clear progression: supervised baselines, physical losses, differentiable solvers, and learned turbulence closures.
Who it is for: Engineers and researchers moving from classical simulation into physics-AI; each chapter ships with executable notebooks.
Tags: TextbookDifferentiable PhysicsPINNNeural OperatorBeginner to Advanced
Related entries
- DeepONet Getting Started · Systematic DeepONet theory and DeepXDE implementation - for PINN developers hand
- Physics Simulation & ML Integration Roadmap · Systematic path from classical PDE numerics to data-driven physics AI - FEM, FVM
- FNO (Fourier Neural Operator) Tutorial · Systematic FNO from theory to code - for engineers and researchers wanting neura
- PINN from Scratch · A classic Chinese structured tutorial implementing PINN from minimal dependencie
- Finite Element from Scratch in Python · Build a 2D linear-elastic FEM solver from NumPy/SciPy - no FEniCS/Abaqus - to de