Improved Depth Estimation of Bayesian Neural Networks

Abstract

This paper proposes improvements over earlier work by Nazareth and Blei (2022) for estimating the depth of Bayesian neural networks. Here, we propose a discrete truncated normal distribution over the network depth to independently learn its mean and variance. Posterior distributions are inferred by minimizing the variational free energy, which balances the model complexity and accuracy. Our method improves test accuracy on the spiral data set and reduces the variance in posterior depth estimates.

PDF NeurIPSW OpenReview Semantic Scholar

Cite

Text

van Erp and de Vries. "Improved Depth Estimation of Bayesian Neural Networks." NeurIPS 2024 Workshops: BDU, 2024.

Markdown

[van Erp and de Vries. "Improved Depth Estimation of Bayesian Neural Networks." NeurIPS 2024 Workshops: BDU, 2024.](https://mlanthology.org/neuripsw/2024/vanerp2024neuripsw-improved/)

BibTeX

@inproceedings{vanerp2024neuripsw-improved,
  title     = {{Improved Depth Estimation of Bayesian Neural Networks}},
  author    = {van Erp, Bart and de Vries, Bert},
  booktitle = {NeurIPS 2024 Workshops: BDU},
  year      = {2024},
  url       = {https://mlanthology.org/neuripsw/2024/vanerp2024neuripsw-improved/}
}