Filters








1,185 Hits in 5.1 sec

Learning for Biomedical Information Extraction: Methodological Review of Recent Advances [article]

Feifan Liu, Jinying Chen, Abhyuday Jagannatha, Hong Yu
2016 arXiv   pre-print
Semi-supervised annotation, active learning, and crowd-sourcing approaches have been developed to create large-scale annotated corpora.  ...  Different from abovementioned semi-supervised methods, distant supervision approach typically doesn't require any labeled data, which makes use of weakly labeled data derived from a knowledge base [97  ... 
arXiv:1606.07993v1 fatcat:7d5om7zxxzhoviiriasrfwg3xi

Semi-supervised Abstraction-Augmented String Kernel for Multi-level Bio-Relation Extraction [chapter]

Pavel Kuksa, Yanjun Qi, Bing Bai, Ronan Collobert, Jason Weston, Vladimir Pavlovic, Xia Ning
2010 Lecture Notes in Computer Science  
In this work we introduce a semisupervised approach that can tackle multi-level bRE via string comparisons with mismatches in the string kernel framework.  ...  This Abstraction-augmented String Kernel (ASK) allows for better generalization of patterns learned from annotated data and provides a unified framework for solving bRE with multiple degrees of detail.  ...  Except self-training and co-training, most of these semi-supervised methods have scalability problems for large scale tasks.  ... 
doi:10.1007/978-3-642-15883-4_9 fatcat:n35jjgwqbncj3jhnf6roebh7yy

Bridging Gap between Image Pixels and Semantics via Supervision: A Survey [article]

Jiali Duan, C.-C. Jay Kuo
2022 arXiv   pre-print
Most importantly, we claim that the semantic gap is primarily bridged through supervised learning today.  ...  Then, it summarizes various supervision methods to bridge the semantic gap in the context of object detection and metric learning.  ...  Pseudo labels are another source of supervision falling into semi-supervised regime. For example, [71] generates pseudo labels based on the pretrained VGG16 features via k-means clustering.  ... 
arXiv:2107.13757v3 fatcat:dw4c74c3h5bvlmzmxugeh5aela

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
Such a setting severely limits their scalability in real-world applications where no labelled samples are available during the training phase.  ...  Most existing person re-identification (Re-ID) approaches follow a supervised learning framework, in which a large number of labelled matching pairs are required for training.  ...  Therefore, these methods can only be considered as semi-supervised or weakly-supervised learning.  ... 
arXiv:1807.01440v2 fatcat:26gw74g7nfblvf53edk32dc4ju

Towards Large-Scale Histopathological Image Analysis: Hashing-Based Image Retrieval

Xiaofan Zhang, Wei Liu, Murat Dundar, Sunil Badve, Shaoting Zhang
2015 IEEE Transactions on Medical Imaging  
Specifically, we present a supervised kernel hashing technique which leverages a small amount of supervised information in learning to compress a 10 000-dimensional image feature vector into only tens  ...  Recently, with the ever-increasing amount of annotated medical data, large-scale and data-driven methods have emerged to offer a promise of bridging the semantic gap between images and diagnostic information  ...  of supervised information (originating from the semantics) that is not well exploited by those unsupervised and semi-supervised methods.  ... 
doi:10.1109/tmi.2014.2361481 pmid:25314696 fatcat:frt22wzviraffm7zjex5wgzlau

Scalable Mobile Video Retrieval with Sparse Projection Learning and Pseudo Label Mining

Guan-Long Wu, Yin-Hsi Kuo, Tzu-Hsuan Chiu, Winston H. Hsu, Lexing Xie
2013 IEEE Multimedia  
To alleviate the need for expensive annotation for hash learning, we investigate varying approaches for pseudo label mining, where explicit semantic analysis leverages Wikipedia and performs the best.  ...  We propose a knowledge-embedded sparse projection learning approach.  ...  To demonstrate the effectiveness of pseudo label mining, Fig. 2 (c) shows the MAP performance in different storage for S3PLH and our proposed algorithm between human annotations (true labels, semi-supervised  ... 
doi:10.1109/mmul.2013.13 fatcat:p7lstaewyvb7nanycftda7vtlq

Transductive Zero-Shot Action Recognition by Word-Vector Embedding [article]

Xun Xu, Timothy Hospedales, Shaogang Gong
2016 arXiv   pre-print
In this work, we explore word-vectors as the shared semantic space to embed videos and category labels for ZSL action recognition.  ...  Existing ZSL studies focus primarily on still images, and attribute-based semantic representations.  ...  Specifically, a KNN graph is constructed in the attribute embedding space and a smoothed solution is obtained transductively by semi-supervised label propagation (Zhou et al, 2004). 5.  ... 
arXiv:1511.04458v2 fatcat:yxfn52pdhjfatedmixz4evtiay

Transductive Zero-Shot Action Recognition by Word-Vector Embedding

Xun Xu, Timothy Hospedales, Shaogang Gong
2017 International Journal of Computer Vision  
The results demonstrate that our approach achieves the state-of-the-art zero-shot action recognition performance with a simple and efficient pipeline, and without supervised annotation of attributes.  ...  Specifically, a KNN graph is constructed in the attribute embedding space and a smoothed solution is obtained transductively by semi-supervised label propagation (Zhou et al, 2004).  ...  Multi-Shot Learning Although our focus is zero-shot learning, we also note that the semantic embedding space provides an alternative representation for conventional supervised learning.  ... 
doi:10.1007/s11263-016-0983-5 fatcat:c6rn4jpg3ff5pbks52ohlcafny

Normalized Cut Loss for Weakly-Supervised CNN Segmentation

Meng Tang, Abdelaziz Djelouah, Federico Perazzi, Yuri Boykov, Christopher Schroers
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Inspired by the general ideas in semi-supervised learning, we address these problems via a new principled loss function evaluating network output with criteria standard in "shallow" segmentation, e.g.  ...  Most recent semantic segmentation methods train deep convolutional neural networks with fully annotated masks requiring pixel-accuracy for good quality training.  ...  Related Work Semi-supervised Learning Semi-supervised learning is about learning from both labeled and unlabeled data [55, 12] .  ... 
doi:10.1109/cvpr.2018.00195 dblp:conf/cvpr/TangDPBS18 fatcat:pv5cejmx5ba4tnqt3wgxc4l55m

Author Index

2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
Kernel Learning Safety in Numbers: Learning Categories from Few Examples with Multi Model Knowledge Transfer Workshop: OM-2: An Online Multi-class Multi-kernel Learning Algorithm Oreifej, Omar  ...  Huang, Yuchi Image Retrieval via Probabilistic Hypergraph Ranking Automatic Image Annotation Using Group Sparsity Huang, Zhongyang Nonparametric Label-to-Region by Search Huber, Martin Lymph Node  ... 
doi:10.1109/cvpr.2010.5539913 fatcat:y6m5knstrzfyfin6jzusc42p54

Semi-supervised Hashing with Semantic Confidence for Large Scale Visual Search

Yingwei Pan, Ting Yao, Houqiang Li, Chong-Wah Ngo, Tao Mei
2015 Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '15  
In addition, the kernelized variant of semi-supervised hashing is also presented.  ...  In this paper, we propose a novel semi-supervised hashing framework by leveraging semantic confidence.  ...  semi-supervised hashing by exploiting pairwise relationship with semantic confidence, ranking semi-supervised hashing by leveraging triplet relationship with semantic confidence and its kernelized variant  ... 
doi:10.1145/2766462.2767725 dblp:conf/sigir/PanYLNM15 fatcat:4p2rycmghjgflcuybwe67fey5y

A Review of Landcover Classification with Very-High Resolution Remotely Sensed Optical Images-Analysis Unit,Model Scalability and Transferability [article]

Rongjun Qin, Tao Liu
2022 arXiv   pre-print
and transferability with a remote sensing classification focus including (1) sparsity and imbalance of data; (2) domain gaps across different geographical regions; and (3) multi-source and multi-view  ...  paper, we present a systematic overview of existing methods by starting from learning methods and varying basic analysis units for landcover mapping tasks, to challenges and solutions on three aspects of scalability  ...  Examples Methods Descriptions Application Scenario in RS Data of Relevant Works Weakly supervised/Se mi-supervised learning Semi-supervised learning aims to address tasks where a small set of labeled data  ... 
arXiv:2202.03342v1 fatcat:wga6z2o2ibhmpatkqroukefjpm

Learning Discriminative Hashing Codes for Cross-Modal Retrieval based on Multi-view Features [article]

Jun Yu, Xiao-Jun Wu, Josef Kittler
2020 arXiv   pre-print
Our framework includes two stages, namely a kernelization process and a quantization process. Kernelization aims to find a common subspace where multi-view features can be fused.  ...  Extensive experiments are performed on single-label datasets (WiKi and MMED) and multi-label datasets (MIRFlickr and NUS-WIDE) and the experimental results indicate the superiority of our method compared  ...  Many unimodal methods, such as Semi-supervised Hashing (SSH) [2] , Semi-supervised manifold-embedded hashing (SMH) [3] , Minimal Loss Hashing (MLH) [4] based on the latent structural SVM framework,  ... 
arXiv:1808.04152v3 fatcat:oio5a7tsifdpzmktx2ecwq44cm

The Emerging Trends of Multi-Label Learning [article]

Weiwei Liu, Xiaobo Shen, Haobo Wang, Ivor W. Tsang
2020 arXiv   pre-print
data with limited supervision to build a multi-label classification model becomes valuable for practical applications, etc.  ...  For example, extreme multi-label classification is an active and rapidly growing research area that deals with classification tasks with an extremely large number of classes or labels; utilizing massive  ...  Semi-Supervised Multi-Label Classification In semi-supervised MLC (SS-MLC) [117] , the dataset is comprised of two sets: fully labeled data and unlabeled data.  ... 
arXiv:2011.11197v2 fatcat:hu6w4vgnwbcqrinrdfytmmjbjm

Recent Advances in Zero-shot Recognition [article]

Yanwei Fu, Tao Xiang, Yu-Gang Jiang, Xiangyang Xue, Leonid Sigal, and Shaogang Gong
2017 arXiv   pre-print
annotated training data.  ...  With the recent renaissance of deep convolution neural networks, encouraging breakthroughs have been achieved on the supervised recognition tasks, where each class has sufficient training data and fully  ...  Other semantic embedding algorithms have also been investigated such as semi-supervised max-margin learning framework [96] , [97] , latent SVM [67] or multi-task learning [20] , [98] , [99] . 3)  ... 
arXiv:1710.04837v1 fatcat:u3mp6dgj2rgqrarjm4dcywegmy
« Previous Showing results 1 — 15 out of 1,185 results