Filters








22,224 Hits in 6.5 sec

Transfer Neural Trees for Heterogeneous Domain Adaptation [chapter]

Wei-Yu Chen, Tzu-Ming Harry Hsu, Yao-Hung Hubert Tsai, Yu-Chiang Frank Wang, Ming-Syan Chen
2016 Lecture Notes in Computer Science  
Heterogeneous domain adaptation (HDA) addresses the task of associating data not only across dissimilar domains but also described by different types of features.  ...  Inspired by the recent advances of neural networks and deep learning, we propose Transfer Neural Trees (TNT) which jointly solves cross-domain feature mapping, adaptation, and classification in a NN-based  ...  In this paper, we propose Transfer Neural Trees (TNT) as a novel NN-based architecture, which can be applied for relating and recognizing heterogeneous cross-domain data.  ... 
doi:10.1007/978-3-319-46454-1_25 fatcat:c47z35dy4nbt3l76cdqqmqmtei

Heterogeneous Domain Adaptation and Classification by Exploiting the Correlation Subspace

Yi-Ren Yeh, Chun-Hao Huang, Yu-Chiang Frank Wang
2014 IEEE Transactions on Image Processing  
We present a novel domain adaptation approach for solving cross-domain pattern recognition problems, i.e., the data or features to be processed and recognized are collected from different domains of interest  ...  each dimension has a unique capability in associating cross-domain data.  ...  Learning Correlation Subspace via CCA or KCCA The idea of domain adaptation for solving cross-domain classification tasks is to determine a common representation (e.g., a joint subspace) for features extracted  ... 
doi:10.1109/tip.2014.2310992 pmid:24710401 fatcat:tqoaukyktvfqja76p2dojesswe

Learning Cross-Domain Landmarks for Heterogeneous Domain Adaptation

Yao-Hung Hubert Tsai, Yi-Ren Yeh, Yu-Chiang Frank Wang
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
While domain adaptation (DA) aims to associate the learning tasks across data domains, heterogeneous domain adaptation (HDA) particularly deals with learning from cross-domain data which are of different  ...  With the goal of deriving a domain-invariant feature subspace for HDA, our CDLS is able to identify representative cross-domain data, including the unlabeled ones in the target domain, for performing adaptation  ...  =1 β i x i u 2 . (5) To match cross-domain conditional data distributions via E C , we apply SVM trained from labeled cross-domain data to predict the pseudo-labels y i u for x i u (as described later  ... 
doi:10.1109/cvpr.2016.549 dblp:conf/cvpr/TsaiYW16 fatcat:3b33il2enzfovdkghngcmhdwue

Recent Advances in Transfer Learning for Cross-Dataset Visual Recognition: A Problem-Oriented Perspective [article]

Jing Zhang and Wanqing Li and Philip Ogunbona and Dong Xu
2019 arXiv   pre-print
Specifically, it categorises the cross-dataset recognition into seventeen problems based on a set of carefully chosen data and label attributes.  ...  This paper takes a problem-oriented perspective and presents a comprehensive review of transfer learning methods, both shallow and deep, for cross-dataset visual recognition.  ...  Cross-modal Recognition The cross-modal transfer, a sub-problem of heterogeneous domain adaptation and heterogeneous transfer learning as shown in Figure 1 , refers to transfer between different data  ... 
arXiv:1705.04396v3 fatcat:iknfmppi5zca7ljovdlwvdwluu

Investigation of Heterogeneity Sources for Occupational Task Recognition via Transfer Learning

Sahand Hajifar, Saeb Ragani Lamooki, Lora A. Cavuoto, Fadel M. Megahed, Hongyue Sun
2021 Sensors  
Our results demonstrated that the support vector machine equipped with domain adaptation outperformed the baseline for cross-sensor, joint cross-subject and cross-sensor, and cross-subject cases, while  ...  This study aims to investigate the impact of four heterogeneity sources, cross-sensor, cross-subject, joint cross-sensor and cross-subject, and cross-scenario heterogeneities, on classification performance  ...  Data Availability Statement: To encourage future research and/or adoption of our work, we have made our MATLAB code available at https://github.com/sahand-hajifar/Occupational-Task-Recognition-via-Domain-Adaptation  ... 
doi:10.3390/s21196677 pmid:34641001 pmcid:PMC8512259 fatcat:7lf5ryt2zbeqfnmey4llcghnku

Recent Advances in Transfer Learning for Cross-Dataset Visual Recognition

Jing Zhang, Wanqing Li, Philip Ogunbona, Dong Xu
2019 ACM Computing Surveys  
Specifically, it categorises the cross-dataset recognition into 17 problems based on a set of carefully chosen data and label attributes.  ...  Specifically, it categorises the cross-dataset recognition into seventeen problems based on a set of carefully chosen data and label attributes.  ...  Figure 4 illustrates the Deep Adaptation Networks (DAN) proposed in [148] ) or joint distributions [151] between domains.  ... 
doi:10.1145/3291124 fatcat:thjzho3xsnfalprmkquldhwpvm

Transfer Adaptation Learning: A Decade Survey [article]

Lei Zhang, Xinbo Gao
2020 arXiv   pre-print
and test data share similar joint probability distribution.  ...  TAL aims to build models that can perform tasks of target domain by learning knowledge from a semantic related but distribution different source domain.  ...  ACKNOWLEDGMENT The author would like to thank the pioneer researchers in transfer learning, domain adaptation and other related fields. The author would also like to thank Dr. Mingsheng Long and Dr.  ... 
arXiv:1903.04687v2 fatcat:wurprqieffalnnp6isfkhh5y5i

Transfer Learning for Speech and Language Processing [article]

Dong Wang, Thomas Fang Zheng
2015 arXiv   pre-print
Transfer learning is closely related to multi-task learning (cross-lingual vs. multilingual), and is traditionally studied in the name of 'model adaptation'.  ...  For example in speech recognition, an acoustic model trained for one language can be used to recognize speech in another language, with little or no re-training data.  ...  Cross-domain transfer learning Cross-domain transfer learning has two different meaning: when the domain refers to applications, then the difference is in the data distribution; when it refers to features  ... 
arXiv:1511.06066v1 fatcat:vzl3rb5oqvauxk3cva6t5r7jzy

Transfer learning for speech and language processing

Dong Wang, Thomas Fang Zheng
2015 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)  
Transfer learning is closely related to multi-task learning (cross-lingual vs. multilingual), and is traditionally studied in the name of 'model adaptation'.  ...  For example in speech recognition, an acoustic model trained for one language can be used to recognize speech in another language, with little or no re-training data.  ...  Cross-domain transfer learning Cross-domain transfer learning has two different meaning: when the domain refers to applications, then the difference is in the data distribution; when it refers to features  ... 
doi:10.1109/apsipa.2015.7415532 dblp:conf/apsipa/WangZ15 fatcat:oby5enn52batdhoewb4n3ufo4y

Adversarial Transfer Learning for Cross-domain Visual Recognition [article]

Shanshan Wang and Lei Zhang and JingRu Fu
2019 arXiv   pre-print
In many practical visual recognition scenarios, feature distribution in the source domain is generally different from that of the target domain, which results in the emergence of general cross-domain visual  ...  With such symmetry of two generators, the input data from source/target domain can be fed into the MLP network for target/source domain generation, supervised by two confrontation oriented coupled discriminators  ...  This network aligned the joint distributions across domains with a joint maximum mean discrepancy (JMMD) criterion. Hu et al.  ... 
arXiv:1711.08904v2 fatcat:7po4z2v2v5gp5nygsywd3htxji

A Survey of Unsupervised Domain Adaptation for Visual Recognition [article]

Youshan Zhang
2021 arXiv   pre-print
While huge volumes of unlabeled data are generated and made available in many domains, the demand for automated understanding of visual data is higher than ever before.  ...  To overcome the burden of annotation, Domain Adaptation (DA) aims to mitigate the domain shift problem when transferring knowledge from one domain into another similar but different domain.  ...  Joint Distribution Adaptation In this setting, many methods minimize the joint distribution Figure 9: The scheme of adaptation regularization based transfer distance between the source domain and  ... 
arXiv:2112.06745v1 fatcat:65ey4xuygrh4fphb5cqwvqi5fq

Domain Agnostic Learning for Unbiased Authentication [article]

Jian Liang, Yuren Cao, Shuang Li, Bing Bai, Hao Li, Fei Wang, Kun Bai
2020 arXiv   pre-print
In our approach, the latent domains are discovered by learning the heterogeneous predictive relationships between inputs and outputs.  ...  Data-driven authentication could be affected by undesired biases, i.e., the models are often trained in one domain (e.g., for people wearing spring outfits) while applied in other domains (e.g., they change  ...  [38] or domain generalization/adaptation [7, 72] .  ... 
arXiv:2010.05250v2 fatcat:h4lrernzuvh2ddxjhvlv7j32si

Deep Visual Domain Adaptation: A Survey [article]

Mei Wang, Weihong Deng
2018 arXiv   pre-print
Deep domain adaption has emerged as a new learning technique to address the lack of massive amounts of labeled data.  ...  First, we present a taxonomy of different deep domain adaption scenarios according to the properties of data that define how two domains are diverged.  ...  To make it more generalized, a joint adaptation network (JAN) [74] aligns the shift in the joint distributions of input features and output labels in multiple domain-specific layers based on a joint  ... 
arXiv:1802.03601v4 fatcat:d5hwwecipjfjzmh7725lmepzfe

Cross-Domain Depression Detection via Harvesting Social Media

Tiancheng Shen, Jia Jia, Guangyao Shen, Fuli Feng, Xiangnan He, Huanbo Luan, Jie Tang, Thanassis Tiropanis, Tat-Seng Chua, Wendy Hall
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
We further propose a cross-domain Deep Neural Network model with Feature Adaptive Transformation & Combination strategy (DNN-FATC) that transfers the relevant information across heterogeneous domains.  ...  In this paper, we study an interesting but challenging problem of enhancing detection in a certain target domain (e.g. Weibo) with ample Twitter data as the source domain.  ...  We proposed a cross-domain Deep Neural Network model with Feature Adaptive Transformation & Combination strategy (DNN-FATC) to transfer the relevant information across heterogeneous domains.  ... 
doi:10.24963/ijcai.2018/223 dblp:conf/ijcai/Shen0SF0L0TCH18 fatcat:u7rewbjgffaatpxcjwlix6xh5m

Domain Adaptation for Visual Applications: A Comprehensive Survey [article]

Gabriela Csurka
2017 arXiv   pre-print
After a general motivation, we first position domain adaptation in the larger transfer learning problem.  ...  domain adaptation methods.  ...  Heterogeneous deepDA. Concerning heterogeneous or multi-modal deep domain adaptation, we can mention the Transfer Neural Trees [156] proposed to relate heterogeneous cross-domain data.  ... 
arXiv:1702.05374v2 fatcat:5va4oz4evjfhxgxddflpbb6pxi
« Previous Showing results 1 — 15 out of 22,224 results