Direct optimization through arg max for discrete variational auto-encoder
Reparameterization of variational auto-encoders with continuous random variables is an effective method for reducing the variance of their gradient estimates. In the discrete case, one can perform reparametrization using the Gumbel-Max trick, but the resulting objective relies on an arg max operatio...
Main Authors: | Gane, Andreea, Jaakkola, Tommi S |
---|---|
Other Authors: | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
Format: | Article |
Language: | English |
Published: |
Morgan Kaufmann Publishers
2021
|
Online Access: | https://hdl.handle.net/1721.1/129438 |
Similar Items
-
Building generative models over discrete structures : from graphical models to deep learning
by: Gane, Georgiana Andreea.
Published: (2019) -
The Variational Homoencoder: Learning to learn high capacity generative models from few examples
by: Hewitt, Luke B., et al.
Published: (2021) -
Disentangled variational auto-encoder for semi-supervised learning
by: Li, Yang, et al.
Published: (2021) -
Max Weber and postmodern theory
by: Gane, Nicholas B.
Published: (1999) -
Variational methods for inference and estimation in graphical models
by: Jaakkola, Tommi S. (Tommi Sakari)
Published: (2005)