Disentangling disentanglement in variational autoencoders
We develop a generalisation of disentanglement in variational autoencoders (VAEs)—decomposition of the latent representation—characterising it as the fulfilment of two factors: a) the latent encodings of the data having an appropriate level of overlap, and b) the aggregate encoding of the data confo...
Huvudupphovsmän: | Mathieu, E, Rainforth, T, Siddharth, N, Teh, Y |
---|---|
Materialtyp: | Conference item |
Publicerad: |
PMLR
2019
|
Liknande verk
-
AD-VAE: Adversarial Disentangling Variational Autoencoder
av: Adson Silva, et al.
Publicerad: (2025-03-01) -
Disentangled Sequential Variational Autoencoder for Collaborative Filtering
av: WU Mei-lin, HUANG Jia-jin, QIN Jin
Publicerad: (2022-12-01) -
Sequential Variational Autoencoder with Adversarial Classifier for Video Disentanglement
av: Takeshi Haga, et al.
Publicerad: (2023-02-01) -
Disentangling Generative Factors of Physical Fields Using Variational Autoencoders
av: Christian Jacobsen, et al.
Publicerad: (2022-06-01) -
On the fairness of disentangled representations
av: Locatello, F, et al.
Publicerad: (2019)