Semi-supervised Disentanglement with Independent Vector Variational Autoencoders

03/14/2020
by   Bo-Kyeong Kim, et al.
8

We aim to separate the generative factors of data into two latent vectors in a variational autoencoder. One vector captures class factors relevant to target classification tasks, while the other vector captures style factors relevant to the remaining information. To learn the discrete class features, we introduce supervision using a small amount of labeled data, which can simply yet effectively reduce the effort required for hyperparameter tuning performed in existing unsupervised methods. Furthermore, we introduce a learning objective to encourage statistical independence between the vectors. We show that (i) this vector independence term exists within the result obtained on decomposing the evidence lower bound with multiple latent vectors, and (ii) encouraging such independence along with reducing the total correlation within the vectors enhances disentanglement performance. Experiments conducted on several image datasets demonstrate that the disentanglement achieved via our method can improve classification performance and generation controllability.

READ FULL TEXT

page 9

page 18

page 19

page 20

page 21

page 22

page 23

page 24

research
08/08/2019

One Model To Rule Them All

We present a new flavor of Variational Autoencoder (VAE) that interpolat...
research
11/03/2015

The Variational Fair Autoencoder

We investigate the problem of learning representations that are invarian...
research
02/01/2021

Semi-Supervised Disentanglement of Class-Related and Class-Independent Factors in VAE

In recent years, extending variational autoencoder's framework to learn ...
research
12/18/2018

A Novel Variational Autoencoder with Applications to Generative Modelling, Classification, and Ordinal Regression

We develop a novel probabilistic generative model based on the variation...
research
09/06/2023

Self-Supervised Disentanglement of Harmonic and Rhythmic Features in Music Audio Signals

The aim of latent variable disentanglement is to infer the multiple info...
research
04/10/2022

Towards efficient representation identification in supervised learning

Humans have a remarkable ability to disentangle complex sensory inputs (...
research
09/12/2018

Hyperprior Induced Unsupervised Disentanglement of Latent Representations

We address the problem of unsupervised disentanglement of latent represe...

Please sign up or login with your details

Forgot password? Click here to reset