$β$-VAEs can retain label information even at high compression
arXiv:1812.02682v16 citations
Originality Synthesis-oriented
AI Analysis
This addresses the problem of unsupervised learning for data compression in machine learning, but it is incremental as it builds on existing β-VAE methods.
The study found that β-VAEs, trained unsupervised, retain significant label information in compressed representations, achieving up to 90% accuracy on Binary Static MNIST and 80% on Omniglot.
In this paper, we investigate the degree to which the encoding of a $β$-VAE captures label information across multiple architectures on Binary Static MNIST and Omniglot. Even though they are trained in a completely unsupervised manner, we demonstrate that a $β$-VAE can retain a large amount of label information, even when asked to learn a highly compressed representation.