Towards Evaluating the Representation Learned by Variational AutoEncoders

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

At the heart of a deep neural network is representation learning with complex latent variables. This representation learning has been improved by disentangled representations and the idea of regularization terms. However, adversarial samples show that tasks with DNNs can easily fail due to slight perturbations or transformations of the input. Variational AutoEncoder (VAE) learns P(z\x), the distribution of the latent variable z, rather than P(y\x), the distribution of the output y for the input x. Therefore, VAE is considered to be a good model for learning representations from input data. In other words, the mapping of x is not directly to y, but to the latent variable z. In this paper, we propose an evaluation method to characterize the latent variables that VAE learns. Specifically, latent variables extracted from VAEs trained by two well-known data sets are analyzed by the k-nearest neighbor method(kNN). In doing so, we propose an interpretation of what kind of representation the VAE learns, and share clues about the hyperdimensional space to which the latent variables are mapped.

Original languageEnglish
Title of host publication2021 60th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages591-594
Number of pages4
ISBN (Electronic)9784907764739
Publication statusPublished - Sep 8 2021
Event60th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2021 - Tokyo, Japan
Duration: Sep 8 2021Sep 10 2021

Publication series

Name2021 60th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2021

Conference

Conference60th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2021
Country/TerritoryJapan
CityTokyo
Period9/8/219/10/21

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Science Applications
  • Control and Optimization
  • Instrumentation

Fingerprint

Dive into the research topics of 'Towards Evaluating the Representation Learned by Variational AutoEncoders'. Together they form a unique fingerprint.

Cite this