DOE CODE   /    Search Results   /    Quasi-Newton Variational Bayes v.3.0

Quasi-Newton Variational Bayes v.3.0

RESOURCE

Publicly Accessible Repository
https://github.com/sandialabs/qnvb
https://doi.org/10.11578/dc.20240826.1

SAVE / SHARE



SNL Badge

Abstract

SAND2024-09052O Quasi-Newton Variational Bayes (QNVB) is a training algorithm that performs high-dimensional variational inference on machine learning models. It provides mechanisms to calibrate and control model uncertainty during training. The PyTorch implementation of projective integral updates for Gaussian mean-fields supports the manuscript, "Projective Integral Updates for High-Dimensional Variational Inference." Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.
Developers:
Duersch, Jed [1][2][3] ;  Safonov, Alexander [1][2][3]
  1. Sandia National Lab. (SNL-CA), Livermore, CA (United States)
  2. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
  3. Sandia National Laboratories (SNL), Albuquerque, NM, and Livermore, CA (United States)
Release Date:
2023-08-09
Project Type:
Open Source, Publicly Available Repository
Software Type:
Scientific
Programming Languages:
Python
PyTorch
Version:
3.0
Licenses:
BSD 2-clause "Simplified" License
Sponsoring Org.:
Code ID:
141481
Site Accession Number:
SCR #2935
Research Org.:
Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
Country of Origin:
United States

RESOURCE

Publicly Accessible Repository
https://github.com/sandialabs/qnvb
https://doi.org/10.11578/dc.20240826.1

SAVE / SHARE



SNL Badge

Citation Formats

Duersch, Jed, and Safonov, Alexander. Quasi-Newton Variational Bayes v.3.0. Computer Software. https://github.com/sandialabs/qnvb. USDOE. 09 Aug. 2023. Web. doi:10.11578/dc.20240826.1.
Duersch, Jed, & Safonov, Alexander. (2023, August 09). Quasi-Newton Variational Bayes v.3.0. [Computer software]. https://github.com/sandialabs/qnvb. https://doi.org/10.11578/dc.20240826.1.
Duersch, Jed, and Safonov, Alexander. "Quasi-Newton Variational Bayes v.3.0." Computer software. August 09, 2023. https://github.com/sandialabs/qnvb. https://doi.org/10.11578/dc.20240826.1.
@misc{ doecode_141481,
title = {Quasi-Newton Variational Bayes v.3.0},
author = {Duersch, Jed and Safonov, Alexander},
abstractNote = {SAND2024-09052O Quasi-Newton Variational Bayes (QNVB) is a training algorithm that performs high-dimensional variational inference on machine learning models. It provides mechanisms to calibrate and control model uncertainty during training. The PyTorch implementation of projective integral updates for Gaussian mean-fields supports the manuscript, "Projective Integral Updates for High-Dimensional Variational Inference." Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.},
doi = {10.11578/dc.20240826.1},
url = {https://doi.org/10.11578/dc.20240826.1},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20240826.1}},
year = {2023},
month = {aug}
}