62,186 Hits in 4.6 sec

Incorporating Multiple Cluster Centers for Multi-Label Learning [article]

Senlin Shu, Fengmao Lv, Yan Yan, Li Li, Shuo He, Jun He
2022 arXiv   pre-print
Multi-label learning deals with the problem that each instance is associated with multiple labels simultaneously.  ...  Most of the existing approaches aim to improve the performance of multi-label learning by exploiting label correlations.  ...  . • IMCC: This is our proposed approach, which incorporates multiple cluster centers for multi-label learning.  ... 
arXiv:2004.08113v3 fatcat:iuoa5wzof5dahlutn3utemcyey

Leveraging loosely-tagged images and inter-object correlations for tag recommendation

Yi Shen, Jianping Fan
2010 Proceedings of the international conference on Multimedia - MM '10  
developed for instance label identification by automatically identifying the correspondences between multiple tags (given at the image level) and the image instances.  ...  To leverage the loosely-tagged images for object classifier training, each loosely-tagged image is partitioned into a set of image instances (image regions) and a multiple instance learning algorithm is  ...  In order to incorporate multi-label images for classifier training, some pioneering work have been done by dividing multi-label learning into a set of binary classification problems or transforming multi-label  ... 
doi:10.1145/1873951.1873956 dblp:conf/mm/ShenF10 fatcat:qhuydkizczhrpjlmbuhmwm26fe

Deep Embedded Multi-view Clustering with Collaborative Training [article]

Jie Xu, Yazhou Ren, Guofeng Li, Lili Pan, Ce Zhu, Zenglin Xu
2020 arXiv   pre-print
A new consistency strategy for cluster centers initialization is further developed to improve the multi-view clustering performance with collaborative training.  ...  Multi-view clustering has attracted increasing attentions recently by utilizing information from multiple views.  ...  For most samples, the predictive soft labels q v ij of their multiple views are consistent and aligned, corresponding to the consensus principle of multi-view clustering.  ... 
arXiv:2007.13067v1 fatcat:l74d2tzhvnadto4ostjrc5ddvi

Center-wise Local Image Mixture For Contrastive Representation Learning [article]

Hao Li, Xiaopeng Zhang, Hongkai Xiong
2021 arXiv   pre-print
This is achieved by searching local similar samples of the anchor, and selecting samples that are closer to the corresponding cluster center, which we denote as center-wise local image selection.  ...  Besides, we introduce multi-resolution augmentation, which enables the representation to be scale invariant. We reach 75.5 ResNet-50, and 59.3  ...  As comparisons, multi-resolution incorporates scale invariance into contrastive learning, and significantly boosts the performance even based on a strong baseline.  ... 
arXiv:2011.02697v3 fatcat:xrplj5rjhbfqvdovju63yhicbi

Deep hierarchical embedding for simultaneous modeling of GPCR proteins in a unified metric space

Taeheon Lee, Sangseon Lee, Minji Kang, Sun Kim
2021 Scientific Reports  
Novel loss term based on metric learning is introduced to incorporate hierarchical relations between proteins. We tested our approach using a public GPCR sequence dataset.  ...  GPCR proteins belong to diverse families of proteins that are defined at multiple hierarchical levels.  ...  Phylogenetic tree in Fig. 3 was generated with scikit-bio 0.5. 6 , an open-source python package for processing bioinformatic data, and GraPhlAn 0.9, an open-source tool for visualizing phylogenetic  ... 
doi:10.1038/s41598-021-88623-8 pmid:33953216 fatcat:cxewfy4ofje5nermpfr6ndmvn4

Semi-supervised document clustering with dual supervision through seeding

Yeming Hu, Evangelos E. Milios, James Blustein
2012 Proceedings of the 27th Annual ACM Symposium on Applied Computing - SAC '12  
Semi-supervised clustering algorithms for general problems use a small amount of labeled instances or pairwise instance constraints to aid the unsupervised clustering.  ...  However, user supervision can also be provided in alternative forms for document clustering, such as labeling a feature by associating it with a document or a cluster.  ...  For each labeled feature w in a document d, it contributes one vote for each of its cluster labels (could be associated with multiple clusters).  ... 
doi:10.1145/2245276.2245306 dblp:conf/sac/HuMB12 fatcat:zr56w4mftrdp7hxougmpzs3vca

Improving EEG Decoding via Clustering-based Multi-task Feature Learning [article]

Yu Zhang, Tao Zhou, Wei Wu, Hua Xie, Hongru Zhu, Guoxu Zhou, Andrzej Cichocki
2020 arXiv   pre-print
To uncover the intrinsic distribution structure of EEG data, we propose a clustering-based multi-task feature learning algorithm for improved EEG pattern decoding.  ...  With the encoded label matrix, we devise a novel multi-task learning algorithm by exploiting the subclass relationship to jointly optimize the EEG pattern features from the uncovered subclasses.  ...  The main contributions of our study can be summarized as follows: (1) We first propose to exploit AP clustering for exploring the underlying structure of EEG data; (2) A clustering-based multi-task learning  ... 
arXiv:2012.06813v1 fatcat:gfiow3fyvvfhhboh46i7hngz5y

Gated recurrent units and temporal convolutional network for multilabel classification [article]

Loris Nanni, Alessandra Lumini, Alessandro Manfe, Riccardo Rampon, Sheryl Brahnam, Giorgio Venturin
2021 arXiv   pre-print
The proposed neural network approach is also combined with Incorporating Multiple Clustering Centers (IMCC), which further boosts classification performance.  ...  Multilabel learning tackles the problem of associating a sample with multiple class labels.  ...  Shu et al., "Incorporating Multiple Cluster Centers for Multi-Label Learning," ArXiv, vol. abs/2004.08113, 2020. [8] M. Ibrahim, M. U. G. Khan, F. Mehmood, M. Asim, and W.  ... 
arXiv:2110.04414v2 fatcat:gyqka2igufhvpau7arqciqtbju

Deep Multi-view Semi-supervised Clustering with Sample Pairwise Constraints [article]

Rui Chen, Yongqiang Tang, Wensheng Zhang, Wenlong Feng
2022 arXiv   pre-print
multi-view clustering loss, semi-supervised pairwise constraint loss and multiple autoencoders reconstruction loss.  ...  Then, we innovatively propose to integrate pairwise constraints into the process of multi-view clustering by enforcing the learned multi-view representation of must-link samples (cannot-link samples) to  ...  Acknowledgment The authors are thankful for the financial support by the National Key Research and Development Program of China (2020AAA0109500), the Key-Area Research and Development Program of Guangdong  ... 
arXiv:2206.04949v1 fatcat:s3vmjwaxznh23hhhzh65oskz3e

A Self-Supervised Framework for Clustering Ensemble [chapter]

Liang Du, Yi-Dong Shen, Zhiyong Shen, Jianying Wang, Zhiwu Xu
2013 Lecture Notes in Computer Science  
Specifically, we treat the base clusterings as pseudo class labels and learn classifiers for each of them.  ...  In this paper, we propose a novel self-supervised learning framework for clustering ensemble.  ...  We would like to thank all anonymous reviewers for their helpful comments. This work is supported in part by NSFC grant 60970045 and China National 973 project 2013CB329305.  ... 
doi:10.1007/978-3-642-38562-9_26 fatcat:rnrhy5iprrayhatmnelegp2tcm

Impact of individual rater style on deep learning uncertainty in medical imaging segmentation [article]

Olivier Vincent, Charley Gros, Julien Cohen-Adad
2021 arXiv   pre-print
The impact of label fusion between raters' annotations on this relationship is also explored, and we show that multi-center consensuses are more effective than single-center consensuses to reduce uncertainty  ...  Two multi-rater public datasets were used, consisting of brain multiple sclerosis lesion and spinal cord grey matter segmentation.  ...  Dice score for consensus Raters average 0.42 Center 1 consensus 0.47 Center 2 consensus 0.46 Center 3 consensus 0.43 Multi-center consensus 0.52 Table 1 : Dice score for different combinations of raters  ... 
arXiv:2105.02197v1 fatcat:aqk3x7pq3vbanls2huychjkmpu

A Multiobjective Simultaneous Learning Framework for Clustering and Classification

Weiling Cai, Songcan Chen, Daoqiang Zhang
2010 IEEE Transactions on Neural Networks  
To overcome that problem, in this paper, we present a multi-objective simultaneous learning framework (named MSCC) for both clustering and classification learning.  ...  on a set of the same parameters, i.e., clustering centers which play a role of the bridge connecting the clustering and classification learning.  ...  BK2008381, National Science Foundation of China under Grant No. 60603029, 60773061 and 60873176, respectively for partial supports.  ... 
doi:10.1109/tnn.2009.2034741 pmid:20028622 fatcat:fzdmqh3dwjhnxfypsvs7s7s4zm

Cluster Ensembles, Majority Vote, Voter Eligibility and Privileged Voters

Masoud Charkhabi, Tarundeep Dhot, Shirin A. Mojarad
2014 International Journal of Machine Learning and Computing  
Contributions of this study include guidance on dealing with the lack of meaningful cluster labels (in the case of ensembles), bimodal cluster distributions and incorporating expert intuition into the  ...  Although Cluster Analysis has become a classical technique for grouping in science and engineering, to the best of our knowledge it's use remains limited in business.  ...  Three challenges are faced when applying cluster ensembles: lack of labels, the need for involving expert intuition and concerns of amplifying error in sequential multi-clustering.  ... 
doi:10.7763/ijmlc.2014.v4.424 fatcat:quknq6qtaje5zedhytegf7ovlq

Unsupervised classification of complex clusters in networks of spiking neurons

S.M. Bohte, J.N. Kok, H. La Poutre
2000 Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium  
Using a temporal Hebbian learning rule, the network architecture yields reliable clustering of high-dimensional multi-modal data.  ...  To correctly classify non-spherical clusters, we present a multi-layer version of the algorithm to perform a form of hierarchical clustering.  ...  Again each data-point is labeled with a marker signifying the winning neuron. (crosses and dots). Figure 4 : 4 Clustering of two interlocking clusters in a multi-layer RBF network.  ... 
doi:10.1109/ijcnn.2000.861316 dblp:conf/ijcnn/BohtePK00 fatcat:mopol6e5bfhybajlmr6eoliyza

Generating superpixels using deep image representations [article]

Thomas Verelst, Matthew Blaschko, Maxim Berman
2019 arXiv   pre-print
A clustering-based superpixel algorithm is transformed into a pixel-wise classification task and superpixel training data is derived from semantic segmentation datasets.  ...  Superpixel algorithms are a common pre-processing step for computer vision algorithms such as segmentation, object tracking and localization.  ...  A multi-label loss could take into account that multiple clusters are good candidates, but we couldn't achieve satisfactory results using this approach.  ... 
arXiv:1903.04586v1 fatcat:vzdwwmko4jckpoq5xtld5j4kpe
« Previous Showing results 1 — 15 out of 62,186 results