-
Multi-source Domain Adaptation for Visual Sentiment Classification
Existing domain adaptation methods on visual sentiment classification ty...
01/12/2020 ∙ by Chuang Lin, et al. ∙
30
∙
share
read it
-
Multi-source Distilling Domain Adaptation
Deep neural networks suffer from performance decay when there is domain ...
11/22/2019 ∙ by Sicheng Zhao, et al. ∙
8
∙
share
read it
-
Curriculum Manager for Source Selection in Multi-Source Domain Adaptation
The performance of Multi-Source Unsupervised Domain Adaptation depends s...
07/02/2020 ∙ by Luyu Yang, et al. ∙
0
∙
share
read it
-
Multi-source Domain Adaptation in the Deep Learning Era: A Systematic Survey
In many practical applications, it is often difficult and expensive to o...
02/26/2020 ∙ by Sicheng Zhao, et al. ∙
38
∙
share
read it
-
Distance Based Source Domain Selection for Sentiment Classification
Automated sentiment classification (SC) on short text fragments has rece...
08/28/2018 ∙ by Lex Razoux Schultz, et al. ∙
0
∙
share
read it
-
Open-Ended Visual Question Answering by Multi-Modal Domain Adaptation
We study the problem of visual question answering (VQA) in images by exp...
11/11/2019 ∙ by Yiming Xu, et al. ∙
0
∙
share
read it
-
Gated Convolutional Neural Networks for Domain Adaptation
Domain Adaptation explores the idea of how to maximize performance on a ...
05/16/2019 ∙ by Avinash Madasu, et al. ∙
0
∙
share
read it
Curriculum CycleGAN for Textual Sentiment Domain Adaptation with Multiple Sources
11/17/2020 ∙ by Sicheng Zhao, et al. ∙
8
∙
share
Sentiment analysis of user-generated reviews or comments on products and services on social media can help enterprises to analyze the feedback from customers and take corresponding actions for improvement. To mitigate large-scale annotations, domain adaptation (DA) provides an alternate solution by learning a transferable model from another labeled source domain. Since the labeled data may be from multiple sources, multi-source domain adaptation (MDA) would be more practical to exploit the complementary information from different domains. Existing MDA methods might fail to extract some discriminative features in the target domain that are related to sentiment, neglect the correlations of different sources as well as the distribution difference among different sub-domains even in the same source, and cannot reflect the varying optimal weighting during different training stages. In this paper, we propose an instance-level multi-source domain adaptation framework, named curriculum cycle-consistent generative adversarial network (C-CycleGAN). Specifically, C-CycleGAN consists of three components: (1) pre-trained text encoder which encodes textual input from different domains into a continuous representation space, (2) intermediate domain generator with curriculum instance-level adaptation which bridges the gap across source and target domains, and (3) task classifier trained on the intermediate domain for final sentiment classification. C-CycleGAN transfers source samples at an instance-level to an intermediate domain that is closer to target domain with sentiment semantics preserved and without losing discriminative features. Further, our dynamic instance-level weighting mechanisms can assign the optimal weights to different source samples in each training stage. We conduct extensive experiments on three benchmark datasets and achieve substantial gains over state-of-the-art approaches.
READ FULL TEXT
Comments
There are no comments yet.