We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: GCVAE: Generalized-Controllable Variational AutoEncoder

Abstract: Variational autoencoders (VAEs) have recently been used for unsupervised disentanglement learning of complex density distributions. Numerous variants exist to encourage disentanglement in latent space while improving reconstruction. However, none have simultaneously managed the trade-off between attaining extremely low reconstruction error and a high disentanglement score. We present a generalized framework to handle this challenge under constrained optimization and demonstrate that it outperforms state-of-the-art existing models as regards disentanglement while balancing reconstruction. We introduce three controllable Lagrangian hyperparameters to control reconstruction loss, KL divergence loss and correlation measure. We prove that maximizing information in the reconstruction network is equivalent to information maximization during amortized inference under reasonable assumptions and constraint relaxation.
Comments: 17 pages, 7 figures, 2 tables
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG)
MSC classes: 62F15, 62F30
Cite as: arXiv:2206.04225 [stat.ML]
  (or arXiv:2206.04225v1 [stat.ML] for this version)

Submission history

From: Kenneth Ezukwoke [view email]
[v1] Thu, 9 Jun 2022 02:29:30 GMT (3356kb,D)

Link back to: arXiv, form interface, contact.