Improving GAN with neighbors embedding and gradient matching

By Ngoc-Trung Tran, Tuan-Anh Bui, Ngai-Man Cheung. 2018

16 citations recorded by the source corpus.

We propose two new techniques for training Generative Adversarial Networks (GANs) in the unsupervised setting. Our objectives are to alleviate mode collapse in GAN and improve the quality of the generated samples. First, we propose neighbor embedding, a manifold learning-based regularization to explicitly retain local structures of latent samples in the generated samples. This prevents generator from producing nearly identical data samples from different latent samples, and reduces mode collapse. We propose an inverse t-SNE regularizer to achieve this. Second, we propose a new technique, gradient matching, to align the distributions of the generated samples and the real samples. As it is challenging to work with high-dimensional sample distributions, we propose to align these distributions through the scalar discriminator scores. We constrain the difference between the discriminator scores of the real samples and generated ones. We further constrain the difference between the gradients of these discriminator scores. We derive these constraints from Taylor approximations of the discriminator function. We perform experiments to demonstrate that our proposed techniques are computationally simple and easy to be incorporated in existing systems. When Gradient matching and Neighbour embedding are applied together, our GN-GAN achieves outstanding results on 1D/2D synthetic, CIFAR-10 and STL-10 datasets, e.g. FID score of 30.80 for the STL-10 dataset. Our code is available at: https://github.com/tntrung/gan

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

InfoMax-GAN: Improved Adversarial Image Generation via Information Maximization and Contrastive Learning (2020, background): “Consequently, recent years have seen efforts [4]–[10] to mitigate these GAN problems, including using gradient matching [5] and a two time-scale update rule [7].”
An Improved Self-supervised GAN via Adversarial Training (2019, methodology): “We use the regularized auto-encoder (AE) in our model to prevent the generator from being severely collapsed and guide the generator in producing samples that resemble real ones as shown in recent works [31, 30].”
On Data Augmentation for GAN Training (2020, methodology): “[35], [36] used auto-encoder to regularize both learning of discriminator and generator.”
Self-supervised GAN: Analysis and Improvement with Multi-class Minimax Game (2019, methodology): “In [42, 43], they combine two different types of supervisory signals: real/fake signals and self-supervised signal in the form of auto-encoder.”