Structure by Architecture: Structured Representations without Regularization
Felix Leeb, Giulia Lanzillotta, Yashas Annadani, Michel Besserve, Stefan Bauer, Bernhard Schölkopf
Abstract
We study the problem of self-supervised structured representation learning using autoencoders for downstream tasks such as generative modeling. Unlike most methods which rely on matching an arbitrary, relatively unstructured, prior distribution for sampling, we propose a sampling technique that relies solely on the independence of latent variables, thereby avoiding the trade-off between reconstruction quality and generative performance typically observed in VAEs. We design a novel autoencoder architecture capable of learning a structured representation without the need for aggressive regularization. Our structural decoders learn a hierarchy of latent variables, thereby ordering the information without any additional regularization or supervision. We demonstrate how these models learn a representation that improves results in a variety of downstream tasks including generation, disentanglement, and extrapolation using several challenging and natural image datasets.
BibTeX
@inproceedings{
leeb2023structure,
title={Structure by Architecture: Structured Representations without Regularization},
author={Felix Leeb and Giulia Lanzillotta and Yashas Annadani and Michel Besserve and Stefan Bauer and Bernhard Sch{\"o}lkopf},
booktitle={The Eleventh International Conference on Learning Representations },
year={2023},
url={https://openreview.net/forum?id=O_lFCPaF48t}
}