Adversarial Learning with Local Coordinate Coding
Adversarial Learning with Local Coordinate Coding
复制标题
DOI：
--
复制
发表时间：
2018-06
期刊：
影响因子：
--
通讯作者：
Jiezhang Cao;Yong Guo;Qingyao Wu;Chunhua Shen;Junzhou Huang;Mingkui Tan
中科院分区：
文献类型：
--
作者：
Jiezhang Cao;Yong Guo;Qingyao Wu;Chunhua Shen;Junzhou Huang;Mingkui Tan
生成对抗网络（GANs）旨在从某些先验分布（例如高斯噪声）中生成逼真的数据。然而，这种先验分布通常与真实数据无关，因此可能会丢失数据的语义信息（例如图像中的几何结构或内容）。在实际应用中，语义信息可能由从数据中学习到的某些潜在分布来表示，然而，这种潜在分布很难在GANs中用于采样。在本文中，我们提出一种基于局部坐标编码（LCC）的采样方法来改进GANs，而不是从预定义的先验分布中采样。我们推导了基于LCC的GANs的泛化界，并证明了较小维度的输入足以实现良好的泛化。在各种真实世界数据集上进行的大量实验证明了所提方法的有效性。
Generative adversarial networks (GANs) aim to generate realistic data from some prior distribution (e.g., Gaussian noises). However, such prior distribution is often independent of real data and thus may lose semantic information (e.g., geometric structure or content in images) of data. In practice, the semantic information might be represented by some latent distribution learned from data, which, however, is hard to be used for sampling in GANs. In this paper, rather than sampling from the pre-defined prior distribution, we propose a Local Coordinate Coding (LCC) based sampling method to improve GANs. We derive a generalization bound for LCC based GANs and prove that a small dimensional input is sufficient to achieve good generalization. Extensive experiments on various real-world datasets demonstrate the effectiveness of the proposed method.