Hyperprior Induced Unsupervised Disentanglement of Latent Representations
This addresses the problem of learning interpretable and independent latent factors for researchers in unsupervised representation learning, though it is incremental as it builds on existing VAE frameworks.
The paper tackles unsupervised disentanglement of latent representations in deep generative models by introducing a hierarchical Bayesian approach with an inverse-Wishart prior on the covariance matrix, achieving better disentanglement and reconstruction than β-VAE and competitive performance with FactorVAE on datasets like 2DShapes and CelebA, with significant gains on a new correlated dataset.
We address the problem of unsupervised disentanglement of latent representations learnt via deep generative models. In contrast to current approaches that operate on the evidence lower bound (ELBO), we argue that statistical independence in the latent space of VAEs can be enforced in a principled hierarchical Bayesian manner. To this effect, we augment the standard VAE with an inverse-Wishart (IW) prior on the covariance matrix of the latent code. By tuning the IW parameters, we are able to encourage (or discourage) independence in the learnt latent dimensions. Extensive experimental results on a range of datasets (2DShapes, 3DChairs, 3DFaces and CelebA) show our approach to outperform the $β$-VAE and is competitive with the state-of-the-art FactorVAE. Our approach achieves significantly better disentanglement and reconstruction on a new dataset (CorrelatedEllipses) which introduces correlations between the factors of variation.