Wasserstein Introspective Neural Networks
This work addresses the need for more efficient and robust neural network models in machine learning, though it is incremental as it builds upon existing INN and WGAN methods.
The paper tackles the problem of enhancing generative modeling and robustness by introducing Wasserstein introspective neural networks (WINN), which integrate generator and discriminator into a single model, achieving nearly a 20 times reduction in model size over introspective neural networks (INN) for unsupervised tasks and improved robustness against adversarial examples in supervised classification.
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.