Generative Adversarial Mapping Networks
This work addresses the challenge of generating high-quality images for computer vision applications, representing an incremental improvement over existing GAN-based methods.
The paper tackled the problem of generating realistic images by improving on Generative Moment Matching Networks (GMMN), which uses maximum mean discrepancy (MMD) but fails on challenging datasets like CIFAR-10 and LSUN. The result is a new model called Generative Adversarial Mapping Networks (GAMN) that adds an adversarial mapper network, significantly outperforming GMMN and achieving superior or comparable performance to state-of-the-art GAN methods on MNIST, CIFAR-10, and LSUN-Bedrooms datasets.
Generative Adversarial Networks (GANs) have shown impressive performance in generating photo-realistic images. They fit generative models by minimizing certain distance measure between the real image distribution and the generated data distribution. Several distance measures have been used, such as Jensen-Shannon divergence, $f$-divergence, and Wasserstein distance, and choosing an appropriate distance measure is very important for training the generative network. In this paper, we choose to use the maximum mean discrepancy (MMD) as the distance metric, which has several nice theoretical guarantees. In fact, generative moment matching network (GMMN) (Li, Swersky, and Zemel 2015) is such a generative model which contains only one generator network $G$ trained by directly minimizing MMD between the real and generated distributions. However, it fails to generate meaningful samples on challenging benchmark datasets, such as CIFAR-10 and LSUN. To improve on GMMN, we propose to add an extra network $F$, called mapper. $F$ maps both real data distribution and generated data distribution from the original data space to a feature representation space $\mathcal{R}$, and it is trained to maximize MMD between the two mapped distributions in $\mathcal{R}$, while the generator $G$ tries to minimize the MMD. We call the new model generative adversarial mapping networks (GAMNs). We demonstrate that the adversarial mapper $F$ can help $G$ to better capture the underlying data distribution. We also show that GAMN significantly outperforms GMMN, and is also superior to or comparable with other state-of-the-art GAN based methods on MNIST, CIFAR-10 and LSUN-Bedrooms datasets.