Latent Binary Bayesian Neural Networks Using 'torch' R package LBBNN version 0.1.5 Latent binary Bayesian neural networks (LBBNNs) are implemented using 'torch', an R interface to the LibTorch backend. Supports mean-field variational inference as well as flexible variational posteriors using normalizing flows. The standard LBBNN implementation follows Hubin and Storvik (2024) <doi:10.3390/math12060788>, using the local reparametrization trick as in Skaaret-Lund et al. (2024) <https://openreview.net/pdf?id=d6kqUKzG3V>. Input-skip connections are also supported, as described in Høyheim et al. (2025) <doi:10.48550/arXiv.2503.10496>. Software Depends: R (>= 3.5) Imports: ggplot2, torch, igraph, coro, svglite Suggests: testthat (>= 3.0.0), knitr, rmarkdown, torchvision Lars Skaaret-Lund <lars.skaaret-lund@nmbu.no> Comprehensive R Archive Network (CRAN) Lars Skaaret-Lund [aut, cre], Aliaksandr Hubin [aut], Eirik Høyheim [aut] MIT + file LICENSE (https://CRAN.R-project.org/package=LBBNN/LICENSE) 2026-04-23 application/tgz https://CRAN.R-project.org/package=LBBNN doi:10.32614/CRAN.package.LBBNN en-US