GILBO: One Metric to Measure Them All

Abstract

We propose a simple, tractable lower bound on the mutual information contained in the joint generative density of any latent variable generative model: the GILBO (Generative Information Lower BOund). It offers a data-independent measure of the complexity of the learned latent variable description, giving the log of the effective description length. It is well-defined for both VAEs and GANs. We compute the GILBO for 800 GANs and VAEs each trained on four datasets (MNIST, FashionMNIST, CIFAR-10 and CelebA) and discuss the results.

Cite

Text

Alemi and Fischer. "GILBO: One Metric to Measure Them All." Neural Information Processing Systems, 2018.

Markdown

[Alemi and Fischer. "GILBO: One Metric to Measure Them All." Neural Information Processing Systems, 2018.](https://mlanthology.org/neurips/2018/alemi2018neurips-gilbo/)

BibTeX

@inproceedings{alemi2018neurips-gilbo,
  title     = {{GILBO: One Metric to Measure Them All}},
  author    = {Alemi, Alexander A and Fischer, Ian},
  booktitle = {Neural Information Processing Systems},
  year      = {2018},
  pages     = {7037-7046},
  url       = {https://mlanthology.org/neurips/2018/alemi2018neurips-gilbo/}
}