Wasserstein Introspective Neural Networks

TL;DR

Wasserstein introspective neural networks (WINN) unify generative and discriminative modeling in a single framework, significantly improving over prior INN methods through Wasserstein distance integration, with benefits in model efficiency and adversarial robustness.

Contribution

WINN introduces a novel model combining generator and discriminator, connecting INN with Wasserstein GANs, and enhances generative and classification performance with fewer parameters.

Findings

Nearly 20 times reduction in model size for unsupervised generative modeling.

Improved robustness against adversarial examples in supervised classification.

Encouraging results on texture, face, and object modeling tasks.

Abstract

We present Wasserstein introspective neural networks (WINN) that are both a generator and a discriminator within a single model. WINN provides a significant improvement over the recent introspective neural networks (INN) method by enhancing INN's generative modeling capability. WINN has three interesting properties: (1) A mathematical connection between the formulation of the INN algorithm and that of Wasserstein generative adversarial networks (WGAN) is made. (2) The explicit adoption of the Wasserstein distance into INN results in a large enhancement to INN, achieving compelling results even with a single classifier --- e.g., providing nearly a 20 times reduction in model size over INN for unsupervised generative modeling. (3) When applied to supervised classification, WINN also gives rise to improved robustness against adversarial examples in terms of the error reduction. In the experiments, we report encouraging results on unsupervised learning problems including texture, face, and object modeling, as well as a supervised classification task against adversarial attacks.