Back
Tags: #ml
- PPCA: The Minimal Generative Model · 8 min read
Deriving probabilistic PCA as the maximum-likelihood solution of a linear-Gaussian latent-variable model, and how it reduces to classical PCA.
- Representing Atomic Environments for ML Potentials · 7 min read
A survey of how local atomic environments are represented in machine-learning interatomic potentials, from Behler-Parrinello symmetry functions to CACE.