arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.14641stat.MLcs.LG
收藏

具有潜在分布学习的最优传输GANs的泛化性质

Generalization Properties of Optimal Transport GANs with Latent Distribution Learning

University College London(伦敦大学学院);Istituto Italiano di Tecnologia(意大利技术研究所);Imperial College London(帝国理工学院)

Giulia Luise, Massimiliano Pontil, Carlo Ciliberto

摘要:

生成式对抗网络(GAN)框架是概率匹配和真实样本生成的一个成熟范例。虽然最近的注意力一直致力于研究这些模型的理论特性,一个完整的理论理解的主要组成部分仍然缺乏。在这项工作中,我们专注于以最优传输指标作为判别器的生成模型,从统计和建模的角度研究了潜在分布和前推映射(生成器)的复杂性之间的相互作用如何影响性能。受我们分析的启发,我们提倡在GAN范式中学习潜在分布以及前推映射。我们证明,这可以导致显着的优势,在样本的复杂性。

英文摘要:

The Generative Adversarial Networks (GAN) framework is a well-established paradigm for probability matching and realistic sample generation. While recent attention has been devoted to studying the theoretical properties of such models, a full theoretical understanding of the main building blocks is still missing. Focusing on generative models with Optimal Transport metrics as discriminators, in this work we study how the interplay between the latent distribution and the complexity of the pushforward map (generator) affects performance, from both statistical and modelling perspectives. Motivated by our analysis, we advocate learning the latent distribution as well as the pushforward map within the GAN paradigm. We prove that this can lead to significant advantages in terms of sample complexity.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑