We introduce effective training algorithms for Generative Adversarial
Networks (GAN) to alleviate mode collapse and gradient vanishing. In our
system, we constrain the generator by an Autoencoder (AE). We propose a
formulation to consider the reconstructed samples from AE as "real" samples for
the discriminator. This couples the convergence of the AE with that of the
discriminator, effectively slowing down the convergence of discriminator and
reducing gradient vanishing. Importantly, we propose two novel distance
constraints to improve the generator. First, we propose a latent-data distance
constraint to enforce compatibility between the latent sample distances and the
corresponding data sample distances. We use this constraint to explicitly
prevent the generator from mode collapse. Second, we propose a
discriminator-score distance constraint to align the distribution of the
generated samples with that of the real samples through the discriminator
score. We use this constraint to guide the generator to synthesize samples that
resemble the real ones. Our proposed GAN using these distance constraints,
namely Dist-GAN, can achieve better results than state-of-the-art methods
across benchmark datasets: synthetic, MNIST, MNIST-1K, CelebA, CIFAR-10 and
STL-10 datasets. Our code is published here (this https URL)
for research.

通过引入自动编码器并提出两个新的距离约束条件，有效地解决了生成对抗网络中的模式崩溃和梯度消失问题，并通过 Dist-GAN 在多个基准数据集上取得了比最先进方法更好的结果。