GILBO: One Metric to Measure Them All

By Alexander A. Alemi, Ian S. Fischer. 2018

18 citations recorded by the source corpus.

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.

Subfields: Generative Models, Evaluation & Benchmarks, Information Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

1. Dan Friedman
2. Adji Bousso Dieng

In Evaluation & Benchmarks

1. Adji Bousso Dieng
2. Dan Friedman

In Information Theory

1. Adji Bousso Dieng
2. Dan Friedman

Citations in the corpus

• The Vendi Score: A Diversity Evaluation Metric for Machine Learning (2022, background): “The Birthday Paradox Test (Arora and Zhang, 2018) aims to estimate the size of the support of a generative model, but requires some manual inspection of samples. gilbo (Alemi and Fischer, 2018) is a reference-free metric but is only applicable to latent variable generative models.”