Part III: Scientific Applications
Applying ML to scientific problems: physics-informed neural networks, symbolic regression, molecular dynamics, and cosmology.
Physics-Informed Neural Networks
PDE constraints, inverse problems, Hamiltonian NNs
Symbolic Regression
Discovering equations from data, genetic programming
ML for Molecular Dynamics
Force fields, equivariant networks, coarse-graining
ML in Cosmology & Astrophysics
Emulators, likelihood-free inference, galaxy classification