Learning Disentanglement in Autoencoders through Euler Encoding

Jaehoon Cha, Jeyan Thiyagalingam

Keywords: disentanglement, disentangling, linear disentangled representations, autoencoder, latent space, factorizing, latent-space factorization, latent-space regularization

TL;DR: We propose the first deterministic model that is aiming to achieve disentanglement based on autoencoders without a pair of images or labels by explicitly introducing inductive biases into a model architecture through Euler encoding.

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 that are independent of one another. To the best of our knowledge, this is the first deterministic model that is aiming to achieve disentanglement based on autoencoders without pairs of images or labels, by explicitly introducing inductive biases into a model architecture through Euler encoding. The proposed model is then compared with a number of state-of-the-art models, relevant to disentanglement, including symmetry-based and generative models based on autoencoders. Our evaluation using six different disentanglement metrics, including the unsupervised disentanglement metric we propose here in this paper, shows that the proposed model can offer better disentanglement, especially when variances of the features are different, where other methods may struggle. We believe that this model opens several opportunities for linear disentangled representation learning based on deterministic autoencoders.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning