Disentangling Autoencoders (DAE)

TL;DR

This paper introduces a novel deterministic autoencoder framework based on symmetry transformations for disentangling latent factors, avoiding regularizers and outperforming existing models in certain scenarios.

Contribution

It presents the first non-probabilistic, regularizer-free autoencoder model for disentanglement using group-theory principles.

Findings

Better disentanglement with varying feature variances

Outperforms seven state-of-the-art autoencoder models

Introduces a new direction for disentanglement learning

Abstract

Noting the importance of factorizing (or disentangling) the latent space, we propose a novel, non-probabilistic disentangling framework for autoencoders, based on the principles of symmetry transformations in group-theory. To the best of our knowledge, this is the first deterministic model that is aiming to achieve disentanglement based on autoencoders without regularizers. The proposed model is compared to seven state-of-the-art generative models based on autoencoders and evaluated based on five supervised disentanglement metrics. The experimental results show that the proposed model can have better disentanglement when variances of each features are different. We believe that this model leads to a new field for disentanglement learning based on autoencoders without regularizers.