Disentangled Representation Learning with Information Maximizing Autoencoder
This work addresses the challenge of unsupervised disentangled representation learning, which is important for improving interpretability and generalization in machine learning, though it appears incremental as it builds on existing autoencoder methods.
The paper tackled the problem of learning disentangled representations from unlabeled data by proposing an Information Maximizing Autoencoder (InfoAE), which achieved 98.9% test accuracy on the MNIST dataset with unsupervised training.
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 ($\pm .1$) $\%$ test accuracy while using complete unsupervised training.