English

Uncertainty in the Variational Information Bottleneck

Machine Learning 2018-07-04 v1 Machine Learning

Abstract

We present a simple case study, demonstrating that Variational Information Bottleneck (VIB) can improve a network's classification calibration as well as its ability to detect out-of-distribution data. Without explicitly being designed to do so, VIB gives two natural metrics for handling and quantifying uncertainty.

Keywords

Cite

@article{arxiv.1807.00906,
  title  = {Uncertainty in the Variational Information Bottleneck},
  author = {Alexander A. Alemi and Ian Fischer and Joshua V. Dillon},
  journal= {arXiv preprint arXiv:1807.00906},
  year   = {2018}
}

Comments

10 pages, 7 figures. Accepted to UAI 2018 - Uncertainty in Deep Learning Workshop

R2 v1 2026-06-23T02:48:45.611Z