VAE learning via Stein variational gradient descent

Yunchen Pu; Zhe Gan; Ricardo Henao; Chunyuan Li; Shaobo Han; Lawrence Carin

VAE learning via Stein variational gradient descent

Yunchen Pu, Zhe Gan, Ricardo Henao, Chunyuan Li, Shaobo Han, Lawrence Carin

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

A new method for learning variational autoencoders (VAEs) is developed, based on Stein variational gradient descent. A key advantage of this approach is that one need not make parametric assumptions about the form of the encoder distribution. Performance is further enhanced by integrating the proposed encoder with importance sampling. Excellent performance is demonstrated across multiple unsupervised and semi-supervised problems, including semi-supervised analysis of the ImageNet data, demonstrating the scalability of the model to large datasets.

Original language	English (US)
Title of host publication	Advances in Neural Information Processing Systems
Publisher	Neural information processing systems foundation
Pages	4237-4246
Number of pages	10
State	Published - Jan 1 2017
Externally published	Yes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2021-02-09

Cite this

@inproceedings{f7451a15f3714547a486972b991eec86,

title = "VAE learning via Stein variational gradient descent",

abstract = "A new method for learning variational autoencoders (VAEs) is developed, based on Stein variational gradient descent. A key advantage of this approach is that one need not make parametric assumptions about the form of the encoder distribution. Performance is further enhanced by integrating the proposed encoder with importance sampling. Excellent performance is demonstrated across multiple unsupervised and semi-supervised problems, including semi-supervised analysis of the ImageNet data, demonstrating the scalability of the model to large datasets.",

author = "Yunchen Pu and Zhe Gan and Ricardo Henao and Chunyuan Li and Shaobo Han and Lawrence Carin",

note = "Generated from Scopus record by KAUST IRTS on 2021-02-09",

year = "2017",

month = jan,

day = "1",

language = "English (US)",

pages = "4237--4246",

booktitle = "Advances in Neural Information Processing Systems",

publisher = "Neural information processing systems foundation",

}

TY - GEN

T1 - VAE learning via Stein variational gradient descent

AU - Pu, Yunchen

AU - Gan, Zhe

AU - Henao, Ricardo

AU - Li, Chunyuan

AU - Han, Shaobo

AU - Carin, Lawrence

N1 - Generated from Scopus record by KAUST IRTS on 2021-02-09

PY - 2017/1/1

Y1 - 2017/1/1

N2 - A new method for learning variational autoencoders (VAEs) is developed, based on Stein variational gradient descent. A key advantage of this approach is that one need not make parametric assumptions about the form of the encoder distribution. Performance is further enhanced by integrating the proposed encoder with importance sampling. Excellent performance is demonstrated across multiple unsupervised and semi-supervised problems, including semi-supervised analysis of the ImageNet data, demonstrating the scalability of the model to large datasets.

AB - A new method for learning variational autoencoders (VAEs) is developed, based on Stein variational gradient descent. A key advantage of this approach is that one need not make parametric assumptions about the form of the encoder distribution. Performance is further enhanced by integrating the proposed encoder with importance sampling. Excellent performance is demonstrated across multiple unsupervised and semi-supervised problems, including semi-supervised analysis of the ImageNet data, demonstrating the scalability of the model to large datasets.

UR - http://www.scopus.com/inward/record.url?scp=85046994842&partnerID=8YFLogxK

M3 - Conference contribution

SP - 4237

EP - 4246

BT - Advances in Neural Information Processing Systems

PB - Neural information processing systems foundation

ER -

VAE learning via Stein variational gradient descent

Abstract

Bibliographical note

Other files and links

Fingerprint

Cite this