Created:
Modified:
Modified:
score matching
This page is from my personal notes, and has not been specifically reviewed for public consumption. It might be incomplete, wrong, outdated, or stupid. Caveat lector.Aapo Hyvärinen: Estimation of Non-Normalized Statistical Models by Score Matching (2005) https://jmlr.org/papers/volume6/hyvarinen05a/hyvarinen05a.pdf
Fitting the score function - the gradient of a log density - rather than the log density itself, allows us to ignore pesky normalization constants.
Todo: develop connections to diffusion models.