Module-based regularization improves Gaussian graphical models when observing noisy data.

Magnus Neuman Joaquín Calatayud Viktor Tasselius Martin Rosvall

Published in: CoRR (2023)

Keyphrases

noisy data
gaussian graphical models
graphical models
input data
missing data
learning from noisy data
linear models
noise tolerant
high dimensional
noise free
high dimensionality
missing values
prior information
belief propagation
pairwise
training data
higher order
least squares
probabilistic model
feature space