Group-based Learning of Disentangled Representations with Generalizability for Novel Contents
This addresses the need for unsupervised methods in representation learning for domains like computer vision, though it is incremental as it builds on existing variational autoencoder frameworks.
The paper tackled the problem of learning disentangled representations from sensory data without explicit labels, achieving high separation between content and transformation factors and generalization to novel contents across five datasets.
Sensory data are often comprised of independent content and transformation factors. For example, face images may have shapes as content and poses as transformation. To infer separately these factors from given data, various ``disentangling'' models have been proposed. However, many of these are supervised or semi-supervised, either requiring attribute labels that are often unavailable or disallowing for generalization over new contents. In this study, we introduce a novel deep generative model, called group-based variational autoencoders. In this, we assume no explicit labels, but a weaker form of structure that groups together data instances having the same content but transformed differently; we thereby separately estimate a group-common factor as content and an instance-specific factor as transformation. This approach allows for learning to represent a general continuous space of contents, which can accommodate unseen contents. Despite the simplicity, our model succeeded in learning, from five datasets, content representations that are highly separate from the transformation representation and generalizable to data with novel contents. We further provide detailed analysis of the latent content code and show insight into how our model obtains the notable transformation invariance and content generalizability.