Disentangling Autoencoders (DAE)
This work addresses the challenge of disentanglement learning for researchers in machine learning, offering a new deterministic approach without regularizers, though it appears incremental as it builds on existing autoencoder methods.
The paper tackles the problem of factorizing latent spaces in autoencoders by proposing a novel, non-probabilistic disentangling framework based on symmetry transformations in group theory, achieving better disentanglement than seven state-of-the-art models when feature variances differ.
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.