Filters








43,339 Hits in 3.0 sec

Unsupervised Domain-Adaptive Person Re-identification Based on Attributes [article]

Xiangping Zhu and Pietro Morerio and Vittorio Murino
2019 arXiv   pre-print
In this work, an unsupervised domain adaptive ReID feature learning framework is proposed to make full use of attribute annotations.  ...  We propose to transfer attribute-related features from their original domain to the ReID one: to this end, we introduce an adversarial discriminative domain adaptation method in order to learn domain invariant  ...  The network is first trained in the attribute recognition domain D a and then adapted to the ReID domain D p . An unsupervised domain adaptation method, based on [13] , is used.  ... 
arXiv:1908.10359v1 fatcat:qlqdce5orbgtjcp5ebyb5nw7wy

Multi-task Mid-level Feature Alignment Network for Unsupervised Cross-Dataset Person Re-Identification [article]

Shan Lin, Haoliang Li, Chang-Tsun Li, Alex Chichung Kot
2018 arXiv   pre-print
To overcome this limitation, we develop a novel unsupervised Multi-task Mid-level Feature Alignment (MMFA) network for the unsupervised cross-dataset person re-identification task.  ...  the attribute learning task with a cross-dataset mid-level feature alignment regularisation term.  ...  The first set of experiments is the unsupervised performance based on the feature representation learned from the source domain attributes or identity, without any domain adaptation.  ... 
arXiv:1807.01440v2 fatcat:26gw74g7nfblvf53edk32dc4ju

Fairness-Aware Node Representation Learning [article]

Öykü Deniz Köse, Yanning Shen
2021 arXiv   pre-print
To this end, this study addresses fairness issues in graph contrastive learning with fairness-aware graph augmentation designs, through adaptive feature masking and edge deletion.  ...  Experimental results on real social networks are presented to demonstrate that the proposed augmentations can enhance fairness in terms of statistical parity and equal opportunity, while providing comparable  ...  It is proved that the adaptive feature masking scheme can reduce the expected correlation between sensitive attributes and nodal features compared with the non-adaptive counterpart, hence reduces the intrinsic  ... 
arXiv:2106.05391v1 fatcat:gbzvlecaeresdjn52ec3osvdwe

Unsupervised Domain Adaptation for Nighttime Aerial Tracking [article]

Junjie Ye, Changhong Fu, Guangze Zheng, Danda Pani Paudel, Guang Chen
2022 arXiv   pre-print
This work instead develops a novel unsupervised domain adaptation framework for nighttime aerial tracking (named UDAT).  ...  Moreover, we construct a pioneering benchmark namely NAT2021 for unsupervised domain adaptive nighttime tracking, which comprises a test set of 180 manually annotated tracking sequences and a train set  ...  Network architecture Feature extractor. Feature extraction of Siamese networks generally consists of two branches, i.e., the template branch and the search branch.  ... 
arXiv:2203.10541v2 fatcat:loc3l4qyjffyzfhhzywniphu2m

Disjoint Label Space Transfer Learning with Common Factorised Space

Xiaobin Chang, Yongxin Yang, Tao Xiang, Timothy M. Hospedales
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
domain adaptation, where the source and target domains share the same label-sets.  ...  It is shared between source and target domains, and trained with an unsupervised factorisation loss and a graph-based loss.  ...  Experiments The proposed model is evaluated on progressively more challenging problems. First, we evaluate CFSM on unsupervised domain adaptation (UDA).  ... 
doi:10.1609/aaai.v33i01.33013288 fatcat:jjnhyseegzginhjhkqsrpykohe

Disjoint Label Space Transfer Learning with Common Factorised Space [article]

Xiaobin Chang, Yongxin Yang, Tao Xiang, Timothy M. Hospedales
2018 arXiv   pre-print
domain adaptation, where the source and target domains share the same label-sets.  ...  It is shared between source and target domains, and trained with an unsupervised factorisation loss and a graph-based loss.  ...  For fair comparison we use an identical feature extractor network to Figure 3 :Figure 4 : 34 (a) is a latent attribute for the colour 'red' cov-CFS activations distribution on target data.  ... 
arXiv:1812.02605v1 fatcat:o26mlpwkbfcxbh5nirdy3fhnt4

Prediction of Thermostability from Amino Acid Attributes by Combination of Clustering with Attribute Weighting: A New Vista in Engineering Enzymes

Mansour Ebrahimi, Amir Lakizadeh, Parisa Agha-Golzadeh, Esmaeil Ebrahimie, Mahdi Ebrahimi, Indra Neil Sarkar
2011 PLoS ONE  
Seventy per cent of the weighting methods selected Gln content and frequency of hydrophilic residues as the most important protein attributes.  ...  Understanding the protein attributes that are involved in this adaptation is the first step toward engineering thermostable enzymes.  ...  One dataset had 794, and the next one had 27 protein features chosen after stepwise feature selection algorithm with various hidden layers in each neural network (Table 3 ).  ... 
doi:10.1371/journal.pone.0023146 pmid:21853079 pmcid:PMC3154288 fatcat:2ba742usbvbbpibm6fp4slo7xy

Machine Learning for Intelligent Authentication in 5G-and-Beyond Wireless Networks [article]

He Fang, Xianbin Wang, Stefano Tomasin
2019 arXiv   pre-print
In this article, we envision new authentication approaches based on machine learning techniques by opportunistically leveraging physical layer attributes, and introduce intelligence to authentication for  ...  and situation-aware device validation under unknown network conditions and unpredictable dynamics.  ...  , network selection in heterogeneous wireless networks, mobility patterns, and communication process.  ... 
arXiv:1907.00429v2 fatcat:qwd4wvenrfhwdcluzzavivdtga

Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-identification

Jingya Wang, Xiatian Zhu, Shaogang Gong, Wei Li
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Specifically, we introduce an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identitydiscriminative feature representation space transferrable  ...  Extensive comparative evaluations validate the superiority of this new TJ-AIDL model for unsupervised person re-id over a wide range of state-of-the-art methods on four challenging benchmarks including  ...  Acknowledgements This work was partially supported by the China Scholarship Council, Vision Semantics Ltd, Royal Society Newton Advanced Fellowship Programme (NA150459), and In-novateUK Industrial Challenge Project on  ... 
doi:10.1109/cvpr.2018.00242 dblp:conf/cvpr/WangZGL18 fatcat:6t46zlmqendb5i2fbuorvsrcw4

Clustering Large Data with Mixed Values Using Extended Fuzzy Adaptive Resonance Theory

Asadi Srinivasulu, Gadupudi Dakshayani
2016 Indonesian Journal of Electrical Engineering and Computer Science  
Finally, the obtained results may consist of clusters which are formed based on the similarity of their attribute type and values.</p>  ...  The previous work deals with unsupervised feature learning techniques such as k-Means and C-Means which cannot be able to process the mixed type of data.  ...  The adaptive resonance theory neural network was used to form the clusters by using this matched prototype, the input vectors are assigned into any one of the clusters based on the similarity of the features  ... 
doi:10.11591/ijeecs.v4.i3.pp617-628 fatcat:og73jx55sfhmvglvr2hrmvvb6a

Deep domain adaptation for describing people based on fine-grained clothing attributes

Qiang Chen, Junshi Huang, Rogerio Feris, Lisa M Brown, Jian Dong, Shuicheng Yan
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
In order to bridge this gap, we propose a novel double-path deep domain adaptation network to model the data from the two domains jointly.  ...  Several alignment cost layers placed inbetween the two columns ensure the consistency of the two domain features and the feasibility to predict unseen attribute categories in one of the domains.  ...  First, selective search is adopted to generate candidate region proposals. Then, a Network-in-Network (NIN) model is used to extract features for each candidate region.  ... 
doi:10.1109/cvpr.2015.7299169 dblp:conf/cvpr/ChenHFBDY15 fatcat:uqxdx75jt5bldd7kruaoovbnoi

Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation [article]

Yen-Cheng Liu, Yu-Ying Yeh, Tzu-Chien Fu, Sheng-De Wang, Wei-Chen Chiu, Yu-Chiang Frank Wang
2018 arXiv   pre-print
While representation learning aims to derive interpretable features for describing visual data, representation disentanglement further results in such features so that particular image attributes can be  ...  Moreover, we further confirm that our model can be applied for solving classification tasks of unsupervised domain adaptation, and performs favorably against state-of-the-art image disentanglement and  ...  In this paper, we propose a novel deep neural networks architecture based on generative adversarial networks (GAN) [9] .  ... 
arXiv:1705.01314v4 fatcat:dtzmrwmuubbvpe6hnotkreg3lm

Prediction of Merchandise Sales on E-Commerce Platforms Based on Data Mining and Deep Learning

Xiaoting Yin, Xiaosha Tao, Rahman Ali
2021 Scientific Programming  
In addition, the experiment concludes that the unsupervised pretrained CNN model is more effective and adaptable in sales forecasting.  ...  evaluating the adaptability of the model in different types of online products.  ...  a model has poor adaptability.  ... 
doi:10.1155/2021/2179692 fatcat:umcj4ha3sfhc3eawettf4cynv4

Evaluation of Rough Set Theory Based Network TrafficData Classifier Using Different Discretization Method

Nandita Sengupta
2012 International Journal of Information and Electronics Engineering  
Different discretization methods are available and selection of one has great impact on classification accuracy, time complexity and system adaptability.  ...  Three discretization methods are applied on continuous KDD network data namely, rough set exploration system (RSES), supervised and unsupervised discretization methods to evaluate the classifier accuracy  ...  In that case some features/attributes may have not been reflected precisely in the data table.  ... 
doi:10.7763/ijiee.2012.v2.110 fatcat:4f57wh3eg5es7bsrelsrkfeuma

Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation

Yen-Cheng Liu, Yu-Ying Yeh, Tzu-Chien Fu, Sheng-De Wang, Wei-Chen Chiu, Yu-Chiang Frank Wang
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
While representation learning aims to derive interpretable features for describing visual data, representation disentanglement further results in such features so that particular image attributes can be  ...  Moreover, we further confirm that our model can be applied for solving classification tasks of unsupervised domain adaptation, and performs favorably against state-of-the-art image disentanglement and  ...  In this paper, we propose a novel deep neural networks architecture based on generative adversarial networks (GAN) [9] .  ... 
doi:10.1109/cvpr.2018.00924 dblp:conf/cvpr/LiuYFWCW18 fatcat:svjk5gwhgnfuzoafmspfoxysey
« Previous Showing results 1 — 15 out of 43,339 results