最小二乘迁移生成对抗网络OA北大核心CSCDCSTPCD
Least Squares Transfer Generative Adversarial Networks
现有的生成对抗网络(Generative Adversarial Networks,GAN)损失函数已经被成功地应用在迁移学习方法中.然而,发现这种损失函数在学习过程中可能会出现梯度消失的问题.为了克服该问题,提出了一种学习领域不变特征的新方法,即最小二乘迁移生成对抗网络(Least Squares Transfer Generative Adversarial Networks,LSTGAN).LSTGAN采用最小二乘生成对抗网络(Least …查看全部>>
The existing Generative Adversarial Networks(GAN)loss function has been successfully used in transfer learning method. However, it is found that this loss function may lead to the vanishing gradients problem during the learn-ing process. To overcome this problem, a new learning domain invariant feature approach, Least Squares Transfer Genera-tive Adversarial Networks(LSTGAN), is proposed. LSTGAN adopts least squares generative adversarial networks loss funct…查看全部>>
王孝顺;陈丹;丘海斌
福州大学 电气工程与自动化学院,福州 350116福州大学 电气工程与自动化学院,福州 350116福州大学 电气工程与自动化学院,福州 350116
信息技术与安全科学
生成对抗网络迁移学习梯度消失领域不变特征最小二乘生成对抗网络损失函数
generative adversarial networkstransfer learningvanishing gradientsdomain invariant featureleast squares generative adversarial networks loss function
《计算机工程与应用》 2019 (14)
24-31,8
福建省自然科学基金(No.2018J01534).
评论