Efficient framework to estimate high-dimensional generalized matrix factorization models using penalized maximum likelihood under a dispersion exponential family specification. Either deterministic and stochastic methods are implemented for the numerical maximization. In particular, the package implements the stochastic gradient descent algorithm with a block-wise mini-batch strategy to speed up the computations and an efficient adaptive learning rate schedule to stabilize the convergence. All the theoretical details can be found in Castiglione et al. (2024, <doi:10.48550/arXiv.2412.20509>). Other methods considered for the optimization are the alternated iterative re-weighted least squares and the quasi-Newton method with diagonal approximation of the Fisher information matrix discussed in Kidzinski et al. (2022, <http://jmlr.org/papers/v23/20-1104.html>).
Version: 1.0.1 Depends: R (≥ 4.0.0), ggplot2 Imports: Rcpp (≥ 1.0.10), RcppArmadillo, RSpectra, parallel, doParallel, foreach, MASS, SuppDists, methods, generics, reshape2, ggpubr, viridisLite LinkingTo: Rcpp, RcppArmadillo Suggests: testthat (≥ 3.0.0), Rtsne, dplyr, knitr, rmarkdown Published: 2025-05-17 DOI: 10.32614/CRAN.package.sgdGMF Author: Cristian Castiglione [aut, cre], Davide Risso [ctb], Alexandre Segers [ctb] Maintainer: Cristian Castiglione <cristian_castiglione at libero.it> BugReports: https://github.com/CristianCastiglione/sgdGMF/issues License: MIT + file LICENSE URL: https://github.com/CristianCastiglione/sgdGMF NeedsCompilation: yes Materials: README NEWS CRAN checks: sgdGMF results Documentation: Downloads: Linking:Please use the canonical form https://CRAN.R-project.org/package=sgdGMF to link to this page.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4