arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2010.03322stat.MLcs.LGmath.OC
收藏

一种在训练生成对抗网络中逃离极限环的方法

A method for escaping limit cycles in training GANs

National Center for Applied Mathematics in Chongqing(重庆国家应用数学中心);Chongqing Normal University(重庆师范大学);School of Mathematical Sciences, University of Electronic Science and Technology of China(电子科技大学数学科学学院)

Li Keke, Yang Xinmin

AI总结:

本文提出预测向心加速算法PCAA以缓解GAN训练中的极限环问题,推导其收敛速率上下界,并结合Adam提出PCAA-Adam,在双线性博弈、多元高斯分布和CelebA数据集上验证了有效性。

摘要:

本文主要开展进一步研究,以通过所提出的预测向心加速算法(PCAA)缓解训练生成对抗网络(GANs)中的极限环行为问题。具体而言,我们首先针对一般双线性博弈推导出PCAA最后迭代收敛速率的上界和下界,其中上界显著改进了先前的结果。然后,我们将PCAA与自适应矩估计算法(Adam)相结合,提出PCAA-Adam,这是一种用于训练GANs的实用方法。最后,我们分别在双线性博弈、多元高斯分布和CelebA数据集上进行了实验,验证了所提出算法的有效性。

英文摘要:

This paper mainly conducts further research to alleviate the issue of limit cycling behavior in training generative adversarial networks (GANs) through the proposed predictive centripetal acceleration algorithm (PCAA). Specifically, we first derive the upper and lower bounds on the last-iterate convergence rates of PCAA for the general bilinear game, with the upper bound notably improving upon previous results. Then, we combine PCAA with the adaptive moment estimation algorithm (Adam) to propose PCAA-Adam, a practical approach for training GANs. Finally, we validate the effectiveness of the proposed algorithm through experiments conducted on bilinear games, multivariate Gaussian distributions, and the CelebA dataset, respectively.

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。

↑