English

Volume-preserving Neural Networks

Machine Learning 2021-04-27 v3 Machine Learning

Abstract

We propose a novel approach to addressing the vanishing (or exploding) gradient problem in deep neural networks. We construct a new architecture for deep neural networks where all layers (except the output layer) of the network are a combination of rotation, permutation, diagonal, and activation sublayers which are all volume preserving. Our approach replaces the standard weight matrix of a neural network with a combination of diagonal, rotational and permutation matrices, all of which are volume-preserving. We introduce a coupled activation function allowing us to preserve volume even in the activation function portion of a neural network layer. This control on the volume forces the gradient (on average) to maintain equilibrium and not explode or vanish. To demonstrate our architecture we apply our volume-preserving neural network model to two standard datasets.

Keywords

Cite

@article{arxiv.1911.09576,
  title  = {Volume-preserving Neural Networks},
  author = {Gordon MacDonald and Andrew Godbout and Bryn Gillcash and Stephanie Cairns},
  journal= {arXiv preprint arXiv:1911.09576},
  year   = {2021}
}

Comments

20 pages, 8 figures

R2 v1 2026-06-23T12:23:34.847Z