论文标题
PIRORGAN：生成对抗网的实际数据
PriorGAN: Real Data Prior for Generative Adversarial Nets
论文作者
论文摘要
生成对抗网络（GAN）在学习丰富的数据分布方面取得了快速的进步。但是，我们争论现有技术中的两个主要问题。首先，学到的分布有大量低质量样本的低质量问题。其次，丢失的模式问题，其中学习分布会错过真实数据分布的某些区域。为了解决这两个问题，我们提出了一份小说，该小说捕获了gan的整个真实数据分布，这称为Priorgans。具体来说，我们采用简单而优雅的高斯混合模型（GMM）来在整个真实数据的功能级别上构建明确的概率分布。通过最大化生成数据的概率，我们可以将低质量样品推向高质量。同时，配备了先验，我们可以估计学习分布中缺少的模式，并在真实数据上设计采样策略以解决问题。提议的实际数据先前可以概括为gan的各种培训设置，例如LSGAN，WGAN-GP，SNGAN甚至StyleGan。我们的实验表明，Priorgans在CIFAR-10，FFHQ，LSUN-CAT和LSUN-BIRD数据集上的最先进。
Generative adversarial networks (GANs) have achieved rapid progress in learning rich data distributions. However, we argue about two main issues in existing techniques. First, the low quality problem where the learned distribution has massive low quality samples. Second, the missing modes problem where the learned distribution misses some certain regions of the real data distribution. To address these two issues, we propose a novel prior that captures the whole real data distribution for GANs, which are called PriorGANs. To be specific, we adopt a simple yet elegant Gaussian Mixture Model (GMM) to build an explicit probability distribution on the feature level for the whole real data. By maximizing the probability of generated data, we can push the low quality samples to high quality. Meanwhile, equipped with the prior, we can estimate the missing modes in the learned distribution and design a sampling strategy on the real data to solve the problem. The proposed real data prior can generalize to various training settings of GANs, such as LSGAN, WGAN-GP, SNGAN, and even the StyleGAN. Our experiments demonstrate that PriorGANs outperform the state-of-the-art on the CIFAR-10, FFHQ, LSUN-cat, and LSUN-bird datasets by large margins.
