PDisVAE replaces the total-correlation penalty with a group-wise partial-correlation penalty, letting VAEs learn independent groups of entangled latent factors.
This also gives us a hint that full disentanglement might be a strong and inappropriate assumption and could result in poor latent interpretation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
A Revisit of Total Correlation in Disentangled Variational Auto-Encoder with Partial Disentanglement
PDisVAE replaces the total-correlation penalty with a group-wise partial-correlation penalty, letting VAEs learn independent groups of entangled latent factors.