(2) 我们设计了一个名为MADAN的新框架来进行语义分割的MDA。除了特征级对齐之外,还通过为每个源循环生成一个自适应域来进一步考虑像素级对齐,该域与新的动态语义一致性损失保持一致。提出了子域聚合鉴别器和跨域循环鉴别器,以更好地对齐不同的自适应域。(3) 我们从合成的GTA和SYNTHIA到真实的Cityscapes和BDDS数据...
Universal multi-source adaptation networkPseudo-margin vectorUnsupervised domain adaptation (DA) enables intelligent models to learn transferable knowledge from a labeled source domain and adapt to a similar but unlabeled target domain. Studies showed that knowledge could be transferred from one source ...
Mutual Learning Network for Multi-Source Domain AdaptationZhenpeng Li 1 , Zhen Zhao 1 , Yuhong Guo 1,2 , Haifeng Shen 1 , Jieping Ye 1AI Tech, DiDi Chuxing, China 1 , Carleton University, Canada 2AbstractEarly Unsupervised Domain Adaptation (UDA) methodshave mostly assumed the setting of ...
In this paper, we study the multi-source heterogeneous domain adaptation problem, and propose a Conditional Weighting Adversarial Network (CWAN) to address it. The proposed CWAN adversarially learns a feature transformer, a label classifier, and a domain discriminator. To quantify the importance of ...
First, a multi-source domain adaptation network is proposed to extract rich transfer features and achieve complementary information from multiple sources. Then, a pseudo-margin vector is employed to handle unseen faults in the target domain and realize the accurate fault diagnosis of RB. Finally, a...
Multi-source domain adaptation involves multiple concurrent task learning, and the gradients are simultaneously back propagated. We validate the proposed framework on multi-source cross-domain sentiment classification datasets in Chinese and English. The experimental results demonstrate that the proposed ...
Recent works of multi-source domain adaptation focus on learning a domain-agnostic model, of which the parameters are static. However, such a static model is difficult to handle conflicts across multiple domains, and suffers from a performance degradation in both source domains and target domain. ...
Multi-source learningMaximum mean discrepancySample reweightingIn recent years, domain adaptation and transfer learning are known as promising techniques with admirable performance to deal with problems with distribution difference between the training (source domain) and test (target domain) data. In this...
align the source and target domains at pixel-level based on Generative Adversarial Network (GAN) [Goodfellow et al., 2014] and its variants, such as CycleGAN [Zhu et al., 2017; Zhao et al., 2019b]. 3. Adversarial discriminative methods employ an adversar- ial objective with a domain ...
Their approach demanded sufficient target data for training the ensemble source network, which may not be practical in many applications. Lee et al.14 introduced a multi-source transfer learning method in image classification, which also addresses the data privacy concerns of the transfer learning ...