Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling

arXiv:2010.13187v2 Announce Type: replace-cross
Abstract: Current autoencoder-based disentangled representation learning methods achieve disentanglement by penalizing the (aggregate) posterior to encourage statistical independence of the latent factors. This approach introduces a trade-off between disentangled representation learning and reconstruction quality since the model does not have enough capacity to learn correlated latent variables that capture detail information present in most image data. To overcome this trade-off, we present a novel multi-stage modeling approach where the disentangled factors are first learned using a …

abstract arxiv autoencoder capacity cs.cv cs.lg current improving learn modeling posterior quality representation representation learning stage statistical stat.ml trade trade-off type via