Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Deep Clustering of Compressed Variational Embeddings

Abstract: 

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.

up
0 users have voted:

Paper Details

Authors:
Suya Wu, Enmao Diao, Jie Ding, Vahid Tarokh
Submitted On:
28 March 2020 - 11:30pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Suya Wu
Session:
Posters
Document Year:
2020
Cite

Document Files

DCC_Deep_Clustering_of_Compressed_Variational_Embeddings_poster.pdf

(44)

Subscribe

[1] Suya Wu, Enmao Diao, Jie Ding, Vahid Tarokh, "Deep Clustering of Compressed Variational Embeddings", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5046. Accessed: Sep. 21, 2020.
@article{5046-20,
url = {http://sigport.org/5046},
author = {Suya Wu; Enmao Diao; Jie Ding; Vahid Tarokh },
publisher = {IEEE SigPort},
title = {Deep Clustering of Compressed Variational Embeddings},
year = {2020} }
TY - EJOUR
T1 - Deep Clustering of Compressed Variational Embeddings
AU - Suya Wu; Enmao Diao; Jie Ding; Vahid Tarokh
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5046
ER -
Suya Wu, Enmao Diao, Jie Ding, Vahid Tarokh. (2020). Deep Clustering of Compressed Variational Embeddings. IEEE SigPort. http://sigport.org/5046
Suya Wu, Enmao Diao, Jie Ding, Vahid Tarokh, 2020. Deep Clustering of Compressed Variational Embeddings. Available at: http://sigport.org/5046.
Suya Wu, Enmao Diao, Jie Ding, Vahid Tarokh. (2020). "Deep Clustering of Compressed Variational Embeddings." Web.
1. Suya Wu, Enmao Diao, Jie Ding, Vahid Tarokh. Deep Clustering of Compressed Variational Embeddings [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5046