Categorical Reparameterization with Gumbel-Softmax

Explore this paper's citation graph

Summary

It is shown that the Gumbel-Softmax estimator outperforms state-of-the-art gradient estimators on structured output prediction and unsupervised generative modeling tasks with categorical latent variables, and enables large speedups on semi-supervised classification.

Type
preprint
Published
2016-11-03
Cited by
6,647
References
33
Access
Open access

Keywords

Categorical variable, Gumbel distribution, Softmax function, Estimator, Latent variable

References

Cited by

Related papers