计算机工程与应用2019，Vol.55Issue(15)：38-46,103,10.DOI:10.3778/j.issn.1002-8331.1812-0225
基于迁移学习的双层生成式对抗网络
Double-Layer Generative Adversarial Networks Based on Transfer Learning
邢恩旭 1吴小勇 1李雅娴2
作者信息
- 1. 北京师范大学 研究生院 珠海分院,珠海市网络与信息安全重点实验室,广东 珠海 519085
- 2. 北京师范大学珠海分校 信息技术学院,广东 珠海 519085
- 折叠
摘要
Abstract
In the confrontation training of the Generative Adversarial Networks(GAN), insufficient training set of target samples will result in the model not being able to accurately learn the corresponding features, but it is difficult to obtain a target sample training set that needs to be manually produced and marked. Aiming at this problem, a two-layer GAN model based on migration learning is proposed. In the first layer network, the pseudo-target samples are used to let the model learn the approximate distribution of the target samples in the structure space, and the model migration is carried out by using the idea of migration learning, and is adjusted according to a small number of target samples in the second layer network. In experiment, the improvement of the model in Chinese font generation and picture frame graph conversion is verified, and a better model is effectively trained in a small number of target sample training sets.关键词
生成式对抗网络/迁移学习/目标样本/字体生成Key words
Generative Adversarial Networks(GAN)/ transfer learning/ target sample/ font generation分类
信息技术与安全科学引用本文复制引用
邢恩旭,吴小勇,李雅娴..基于迁移学习的双层生成式对抗网络[J].计算机工程与应用,2019,55(15):38-46,103,10.
