Towards Conceptual Compression

TL;DR

This paper presents a recurrent variational auto-encoder that advances image modeling by effectively capturing global concepts and details, enabling high-quality conceptual compression and setting new state-of-the-art results on ImageNet and Omniglot datasets.

Contribution

It introduces a simple recurrent variational auto-encoder that separates global concepts from details, improving image modeling and enabling conceptual compression.

Findings

Achieves state-of-the-art results on ImageNet and Omniglot datasets.

Effectively separates global conceptual information from details.

Enables high-quality conceptual compression.

Abstract

We introduce a simple recurrent variational auto-encoder architecture that significantly improves image modeling. The system represents the state-of-the-art in latent variable models for both the ImageNet and Omniglot datasets. We show that it naturally separates global conceptual information from lower level details, thus addressing one of the fundamentally desired properties of unsupervised learning. Furthermore, the possibility of restricting ourselves to storing only global information about an image allows us to achieve high quality 'conceptual compression'.