arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.02664cs.LGcs.CGstat.ML
收藏

几何分数:一种比较生成对抗网络的方法

Geometry Score: A Method For Comparing Generative Adversarial Networks

Skolkovo Institute of Science and Technology(斯科尔科沃科学技术研究院);Institute of Numerical Mathematics RAS(俄罗斯科学院数值数学研究所)

Valentin Khrulkov, Ivan Oseledets

摘要:

生成对抗网络(GANs)研究中的最大挑战之一是评估生成样本的质量并检测各种级别的模式坍塌。在这项工作中,我们构建了一个新的衡量性能的GAN的基础数据流形和生成的,这提供了定性和定量的评价手段的几何属性进行比较。我们的算法可以应用于任意性质的数据集,并不限于视觉数据。我们在各种现实模型和数据集上测试了所获得的度量,并证明我们的方法为GANs的属性提供了新的见解。

英文摘要:

One of the biggest challenges in the research of generative adversarial networks (GANs) is assessing the quality of generated samples and detecting various levels of mode collapse. In this work, we construct a novel measure of performance of a GAN by comparing geometrical properties of the underlying data manifold and the generated one, which provides both qualitative and quantitative means for evaluation. Our algorithm can be applied to datasets of an arbitrary nature and is not limited to visual data. We test the obtained metric on various real-life models and datasets and demonstrate that our method provides new insights into properties of GANs.

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。

↑