5,522 Hits in 5.4 sec

Active Learning for Probabilistic Structured Prediction of Cuts and Matchings

Sima Behpour, Anqi Liu, Brian D. Ziebart
2019 International Conference on Machine Learning  
We propose an adversarial approach for active learning with structured prediction domains that is tractable for cuts and matching.  ...  Specifically, while non-probabilistic methods based on structured support vector machines can be tractably applied to predicting cuts and bipartite matchings, conditional random fields are intractable  ...  We employ pool-based active learning (Lewis & Gale, 1994) for the bipartite matching problem using Adversarial Bipartite Matching (ABM), as shown in Algorithm 1.  ... 
dblp:conf/icml/BehpourLZ19 fatcat:3kienvxtwnfflgmlpa7mtuz744

Active Learning in Video Tracking [article]

Sima Behpour
2020 arXiv   pre-print
Specifically, while non-probabilistic methods based on structured support vector machines can be tractably applied to predicting bipartite matchings, conditional random fields are intractable for these  ...  We propose an adversarial approach for active learning with structured prediction domains that is tractable for matching.  ...  Acknowledgement I would like to thank Brian Ziebart for all his advise and support on this work.  ... 
arXiv:1912.12557v3 fatcat:sznqd3qrrbdwbpwt3alswbxpl4

A Survey of Recent View-based 3D Model Retrieval Methods [article]

Qiong Liu
2012 arXiv   pre-print
For matching between multiple views, the many-to-many matching, probabilistic matching and semisupervised learning methods are introduced.  ...  Recently, view-based methods have attracted much research attention due to the high discriminative property of multi-views for 3D object representation.  ...  The contour-based feature was extracted for each view for multi-view matching.  ... 
arXiv:1208.3670v1 fatcat:liljkz7ilrdqbeua62qmzynx3y

Restricted Bipartite Graphs Based Target Detection for Hyperspectral Image Classification with GFA-LFDA Multi Feature Selection

T. Karthikeyan, S. Venkatesh Kumar
2015 Research Journal of Applied Sciences Engineering and Technology  
Proposed system uses a graph based representation, Restricted Bipartite Graphs (RBG) for exact detection of the class values.  ...  Before that the feature of the HSI images are selected using the Gaussian Firefly Algorithm (GFA) for multiple feature selection and Local-Fisher's Discriminant Analysis (LFDA) based feature projection  ...  This work focus a segmentation method based on the multi-scale grid for DMP concept.  ... 
doi:10.19026/rjaset.10.2457 fatcat:mp4fjjk3czdvhavxwz5bonphtu

A System for Multi-label Classification of Learning Objects [chapter]

Vivian F. López Batista, Fernando Prieta Pintado, Ana Belén Gil, Sara Rodríguez, María N. Moreno
2011 Advances in Intelligent and Soft Computing  
For this classification, it is used a special multi-label data mining designed for the LO ranking tasks.  ...  The learning process is supervised, using two major tasks in supervised learning from multi-label data: multi-label classification and label ranking.  ...  selecting learning materials establishing a ranking system for the LOs.  ... 
doi:10.1007/978-3-642-19644-7_55 dblp:conf/softcomp/BatistaPGRG11 fatcat:7qutd3miqrh55antc4jjfk6fke

Findings on Conversation Disentanglement [article]

Rongxin Zhu, Jey Han Lau, Jianzhong Qi
2021 arXiv   pre-print
Observing that the ground truth label (past utterance) is in the top candidates when our model makes an error, we experiment with using bipartite graphs as a post-processing step to learn how to best match  ...  We then build a multi-task learning model that jointly learns utterance-to-utterance and utterance-to-thread classification.  ...  Acknowledgement We would like to thank the anonymous reviewers for their helpful comments.  ... 
arXiv:2112.05346v1 fatcat:dell6quuc5gv3mtnjjotdt6cv4

End-to-End Bootstrapping Neural Network for Entity Set Expansion

Lingyong Yan, Xianpei Han, Ben He, Le Sun
Bootstrapping for entity set expansion (ESE) has long been modeled as a multi-step pipelined process.  ...  problem; 2) it is hard to exploit the high-order entity-pattern relations for entity set expansion.  ...  To learn our BootstrapNet, we devise a multi-view based learning algorithm, which can efficiently learn our model using a small set of seed entities.  ... 
doi:10.1609/aaai.v34i05.6482 fatcat:5traskqpq5ev5m4ndrjngivhwq

SCIM: universal single-cell matching with unpaired feature sets

Stefan G Stark, Joanna Ficek, Francesco Locatello, Ximena Bonilla, Stéphane Chevrier, Franziska Singer, Rudolf Aebersold, Faisal S Al-Quaddoomi, Jonas Albinus, Ilaria Alborelli, Sonali Andani, Per-Olof Attinger (+115 others)
2020 Bioinformatics  
Multi-modal datasets are integrated by pairing cells across technologies using a bipartite matching scheme that operates on the low-dimensional latent representations.  ...  90% and 78% cell-matching accuracy for each one of the samples, respectively.  ...  F.L. was supported by the Max Planck ETH Center for Learning Systems, by an ETH core grant (to G.R.) and by a Google Ph.D. Fellowship. F.L. contributed to this work while working at ETH Zurich.  ... 
doi:10.1093/bioinformatics/btaa843 pmid:33381818 fatcat:mysixyzzjbcjphtutpkgn5ak7i

PRISM: Person Re-Identification via Structured Matching [article]

Ziming Zhang, Venkatesh Saligrama
2015 arXiv   pre-print
We view the global problem as a weighted graph matching problem and estimate edge weights by learning to predict them based on the co-occurrences of visual patterns in the training examples.  ...  We propose PRISM, a structured matching method to jointly account for these challenges.  ...  CONCLUSION In this paper, we propose a structured matching based method for re-id in the contexts of (1) single-shot learning, and (2) multi-shot learning.  ... 
arXiv:1406.4444v4 fatcat:qu2e4fskkfgtzgwwxbs7cei7va

Identifying Protein Complexes in Protein-protein Interaction Data using Graph Convolution Network [article]

Nazar Zaki, Harsh Singh
2021 bioRxiv   pre-print
A representation learning approach, which combines the multi-class GCN feature extractor (to obtain the features of the nodes) and the mean shift clustering algorithm (to perform clustering), is also presented  ...  (MCLA) and Hybrid Bipartite Graph Formulation (HBGF) algorithm.  ...  The predicted outcome for a sample is called predicted matched only if both the outcome and label have one at the same place.  ... 
doi:10.1101/2021.07.07.451457 fatcat:c5g6iiqrtzad7nx6wqn242ihdi

A multi-label classification method using a hierarchical and transparent representation for paper-reviewer recommendation [article]

Dong Zhang, Shu Zhao, Zhen Duan, Jie Chen, Yangping Zhang, Jie Tang
2019 arXiv   pre-print
Further, we propose a simple multi-label-based reviewer assignment MLBRA strategy to select the appropriate reviewers.  ...  In this paper, we propose a Multi-Label Classification method using a hierarchical and transparent Representation named Hiepar-MLC.  ...  Natural Science Foundation of China (Grants #61876001, #61602003 and #61673020), the Provincial Natural Science Foundation of Anhui Province (#1708085QF156), and the Recruitment Project of Anhui University for  ... 
arXiv:1912.08976v1 fatcat:c2sb7yupjrgz3alzfudcflup2y

Transfer re-identification: From person to set-based verification

Wei-Shi Zheng, Shaogang Gong, Tao Xiang
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
To this end, we formulate a transfer learning framework for mining discriminant information from non-target people data to solve the watch list set verification problem.  ...  We also design new criteria for evaluating the performance of the proposed transfer learning method against the i-LIDS and ETHZ data sets.  ...  A learned statistical models for matching two images of a person captured at two different camera views can be easily overfitted. Set-based Verification.  ... 
doi:10.1109/cvpr.2012.6247985 dblp:conf/cvpr/ZhengGX12 fatcat:j7soauzqjzhrzj5rb2qp6pbnoe

Online Multi-target Tracking by Large Margin Structured Learning [chapter]

Suna Kim, Suha Kwak, Jan Feyereisl, Bohyung Han
2013 Lecture Notes in Computer Science  
This problem is formulated as a bipartite matching and solved by a generalized classification, specifically, Structural Support Vector Machines (S-SVM).  ...  Our structural classifier is trained based on matching results given the similarities between all pairs of objects identified in two consecutive frames, where the similarity can be defined by various features  ...  Overview of our multi-target tracking framework by structural prediction. We learn the matching structure and the inference for testing is optimized by the Hungarian algorithm in a bipartite graph.  ... 
doi:10.1007/978-3-642-37431-9_8 fatcat:izvf2wm4s5c37grwdh3p5sclli

Compound Prototype Matching for Few-shot Action Recognition [article]

Lijin Yang, Yifei Huang, Yoichi Sato
2022 arXiv   pre-print
For the focused prototypes, since actions have various temporal variations in the videos, we apply bipartite matching to allow the comparison of actions with different temporal positions and shifts.  ...  Few-shot action recognition aims to recognize novel action classes using only a small number of labeled training samples.  ...  For the focused prototypes P f , we apply a bipartite matching-based similarity measure.  ... 
arXiv:2207.05515v3 fatcat:6t242zsjzbffjozctxidqwgfn4

Unsupervised Feature Learning with Graph Embedding for View-based 3D Model Retrieval

Yu-Ting Su, Wen-Hui Li, Wei-Zhi Nie, Dan Song, An-An Liu
2019 IEEE Access  
However, existing approaches usually learn discriminative visual features and develop a complex graph matching strategy to measure the similarity independently.  ...  For this challenging task, feature learning and similarity measure are two critical problems.  ...  ) [22] and Camera Constraint Free View (CCFV) [25] , and three graph-based methods, Weighted Bipartite Graph Matching (WBGM) [24] , Multi-Modal Clique-Graph Matching (MCG) [26] and Hierarchical Graph  ... 
doi:10.1109/access.2019.2929109 fatcat:cl4eb6ullfghjk7gmgsxgkafny
« Previous Showing results 1 — 15 out of 5,522 results