An autoencoder trained with a commutative swap loss separates color and shape into distinct latent subspaces without feature labels.
Higgins, et al., β -vae: Learning basic visual concepts with a con- strained variational framework, Proc
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
unclear 1representative citing papers
citing papers explorer
-
Unsupervised categorization of similarity measures
An autoencoder trained with a commutative swap loss separates color and shape into distinct latent subspaces without feature labels.