Learning disentangled representation from any unlabelled data is a non-trivial problem. In this paper we propose Information Maximising Autoencoder (InfoAE) where the encoder learns powerful disentangled representation through maximizing the mutual information between the representation and given information in an unsupervised fashion. We have evaluated our model on MNIST dataset and achieved 98.9 (±.1) % test accuracy while using complete unsupervised training.
@article{arxiv.1904.08613,
title = {Disentangled Representation Learning with Information Maximizing Autoencoder},
author = {Kazi Nazmul Haque and Siddique Latif and Rajib Rana},
journal= {arXiv preprint arXiv:1904.08613},
year = {2019}
}