Corpus ID: 209202551

A Closer Look at Disentangling in $\beta$-VAE

  title={A Closer Look at Disentangling in \$\beta\$-VAE},
  author={Harshvardhan Sikka and Weishun Zhong and J. Yin and Cengiz Pehlevan},
  journal={arXiv: Machine Learning},
  • Harshvardhan Sikka, Weishun Zhong, +1 author Cengiz Pehlevan
  • Published 2019
  • Mathematics, Computer Science
  • arXiv: Machine Learning
  • In many data analysis tasks, it is beneficial to learn representations where each dimension is statistically independent and thus disentangled from the others. If data generating factors are also statistically independent, disentangled representations can be formed by Bayesian inference of latent variables. We examine a generalization of the Variational Autoencoder (VAE), $\beta$-VAE, for learning such representations using variational inference. $\beta$-VAE enforces conditional independence of… CONTINUE READING

    Figures, Tables, and Topics from this paper.


    Publications referenced by this paper.
    Understanding disentangling in $\beta$-VAE
    • 157
    • PDF
    Understanding disentangling in β-VAE
    • 186
    Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
    • 279
    • PDF
    Learning Disentangled Representations with Semi-Supervised Deep Generative Models
    • 174
    • PDF
    An Information-Theoretic Analysis of Deep Latent-Variable Models
    • 56
    • PDF
    Variational Autoencoders and Nonlinear ICA: A Unifying Framework
    • 49
    • PDF
    Towards a Definition of Disentangled Representations
    • 92
    • PDF
    Auto-Encoding Variational Bayes
    • 9,148
    • Highly Influential
    • PDF
    Tutorial on Variational Autoencoders
    • 632
    • PDF