Institute Homepage DE Sign In

Back

Empirical Inference Conference Paper 2021

An Infinite-Feature Extension for Bayesian ReLU Nets That Fixes Their Asymptotic Overconfidence

Thumb ticker sm hennig lowres cropped
Probabilistic Numerics, Empirical Inference
Philipp Hennig
Affiliated Researcher

Author(s): Kristiadi, A. and Hein, M. and Hennig, P.
Book Title: Advances in Neural Information Processing Systems 34 (NeurIPS 2021)
Pages: 18789--18800
Year: 2021
Month: December
Editors: M. Ranzato and A. Beygelzimer and Y. Dauphin and P.S. Liang and J. Wortman Vaughan
Publisher: Curran Associates, Inc.
Bibtex Type: Conference Paper (conference)
Event Name: 35th Annual Conference on Neural Information Processing Systems
Event Place: Virtual Conference
State: Published
URL: https://proceedings.neurips.cc/paper/2021/file/9be40cee5b0eee1462c82c6964087ff9-Paper.pdf
Electronic Archiving: grant_archive

BibTex 

@conference{KriHeiHen21b,
  title = {An Infinite-Feature Extension for Bayesian ReLU Nets That Fixes Their Asymptotic Overconfidence},
  booktitle = {Advances in Neural Information Processing Systems 34 (NeurIPS 2021)},
  pages = {18789--18800},
  editors = {M. Ranzato and A. Beygelzimer and Y. Dauphin and P.S. Liang and J. Wortman Vaughan},
  publisher = {Curran Associates, Inc.},
  month = dec,
  year = {2021},
  slug = {kriheihen21b},
  author = {Kristiadi, A. and Hein, M. and Hennig, P.},
  url = {https://proceedings.neurips.cc/paper/2021/file/9be40cee5b0eee1462c82c6964087ff9-Paper.pdf},
  month_numeric = {12}
}