Hierarchical Adversarially Learned Inference
This work addresses the challenge of discovering semantically meaningful hierarchical structures in data for researchers in unsupervised and semi-supervised learning, though it is incremental in combining hierarchical models with adversarial methods.
The paper tackles the problem of learning hierarchical latent representations in an unsupervised manner using adversarial training, achieving competitive performance on attribute prediction with CelebA and state-of-the-art results on semi-supervised MNIST classification.
We propose a novel hierarchical generative model with a simple Markovian structure and a corresponding inference model. Both the generative and inference model are trained using the adversarial learning paradigm. We demonstrate that the hierarchical structure supports the learning of progressively more abstract representations as well as providing semantically meaningful reconstructions with different levels of fidelity. Furthermore, we show that minimizing the Jensen-Shanon divergence between the generative and inference network is enough to minimize the reconstruction error. The resulting semantically meaningful hierarchical latent structure discovery is exemplified on the CelebA dataset. There, we show that the features learned by our model in an unsupervised way outperform the best handcrafted features. Furthermore, the extracted features remain competitive when compared to several recent deep supervised approaches on an attribute prediction task on CelebA. Finally, we leverage the model's inference network to achieve state-of-the-art performance on a semi-supervised variant of the MNIST digit classification task.