Skip to main content

Deep clustering of compressed variational embeddings

Publication ,  Journal Article
Wu, S; Diao, E; Ding, J; Tarokh, V
Published in: Data Compression Conference Proceedings
March 1, 2020

Motivated by the ever-increasing demands for limited communication bandwidth and low-power consumption, we propose a new methodology, named joint Variational Autoencoders with Bernoulli mixture models (VAB), for performing clustering in the compressed data domain. The idea is to reduce the data dimension by Variational Autoencoders (VAEs) and group data representations by Bernoulli mixture models (BMMs). Once jointly trained for compression and clustering, the model can be decomposed into two parts: a data vendor that encodes the raw data into compressed data, and a data consumer that classifies the received (compressed) data. In this way, the data vendor benefits from data security and communication bandwidth, while the data consumer benefits from low computational complexity. To enable training using the gradient descent algorithm, we propose to use the Gumbel-Softmax distribution to resolve the infeasibility of the back-propagation algorithm when assessing categorical samples.

Duke Scholars

Altmetric Attention Stats
Dimensions Citation Stats

Published In

Data Compression Conference Proceedings

DOI

ISSN

1068-0314

Publication Date

March 1, 2020

Volume

2020-March

Start / End Page

399
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Wu, S., Diao, E., Ding, J., & Tarokh, V. (2020). Deep clustering of compressed variational embeddings. Data Compression Conference Proceedings, 2020-March, 399. https://doi.org/10.1109/DCC47342.2020.00051
Wu, S., E. Diao, J. Ding, and V. Tarokh. “Deep clustering of compressed variational embeddings.” Data Compression Conference Proceedings 2020-March (March 1, 2020): 399. https://doi.org/10.1109/DCC47342.2020.00051.
Wu S, Diao E, Ding J, Tarokh V. Deep clustering of compressed variational embeddings. Data Compression Conference Proceedings. 2020 Mar 1;2020-March:399.
Wu, S., et al. “Deep clustering of compressed variational embeddings.” Data Compression Conference Proceedings, vol. 2020-March, Mar. 2020, p. 399. Scopus, doi:10.1109/DCC47342.2020.00051.
Wu S, Diao E, Ding J, Tarokh V. Deep clustering of compressed variational embeddings. Data Compression Conference Proceedings. 2020 Mar 1;2020-March:399.

Published In

Data Compression Conference Proceedings

DOI

ISSN

1068-0314

Publication Date

March 1, 2020

Volume

2020-March

Start / End Page

399