Filters








1,049 Hits in 4.0 sec

Generalized Multiview Analysis: A discriminative latent space

A. Sharma, A. Kumar, H. Daume, D. W. Jacobs
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
This paper presents a general multi-view feature extraction approach that we call Generalized Multiview Analysis or GMA.  ...  GMA solves a joint, relaxed QCQP over different feature spaces to obtain a single (non)linear subspace.  ...  Generalized Multiview Analysis We now present a generalization of this framework to a multi-view setting.  ... 
doi:10.1109/cvpr.2012.6247923 dblp:conf/cvpr/SharmaKDJ12 fatcat:aeqgsz62hfesbjwq4pikwr6tj4

Biconditional Generative Adversarial Networks for Multiview Learning with Missing Views [article]

Anastasiia Doinychko, Massih-Reza Amini
2019 arXiv   pre-print
Our proposed approach addresses this problem by jointly learning the missing views and the multiview classifier using a tripartite game with two generators and a discriminator.  ...  In this paper, we present a conditional GAN with two generators and a common discriminator for multiview learning problems where observations have two views, but one of them may be missing for some of  ...  The former approaches rely mostly on an encoder-encoder network to first map images into a latent space and then generate their views using an inverse mapping.  ... 
arXiv:1911.01861v2 fatcat:nyfsni7kcre6jpyzuzxozxxjua

Generative Models as a Data Source for Multiview Representation Learning [article]

Ali Jahanian, Xavier Puig, Yonglong Tian, Phillip Isola
2022 arXiv   pre-print
We show that for contrastive methods, this multiview data can naturally be used to identify positive pairs (nearby in latent space) and negative pairs (far apart in latent space).  ...  We compare several representation learning methods that can be applied to this setting, using the latent space of the generator to generate multiple "views" of the same semantic content.  ...  (b) With a generative model, we can instead create views by sampling nearby points in latent space Z, exploiting the fact that nearby points in latent space tend to generate imagery of the same semantic  ... 
arXiv:2106.05258v3 fatcat:axi62adbwzcmbbgm6l4kxcybyq

Multiview LSA: Representation Learning via Generalized CCA

Pushpendre Rastogi, Benjamin Van Durme, Raman Arora
2015 Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies  
Multiview LSA (MVLSA) is a generalization of Latent Semantic Analysis (LSA) that supports the fusion of arbitrary views of data and relies on Generalized Canonical Correlation Analysis (GCCA).  ...  Experiments across a comprehensive collection of test-sets show our approach to be competitive with the state of the art.  ...  One of the earliest linguistic vector space models was Latent Semantic Analysis (LSA).  ... 
doi:10.3115/v1/n15-1058 dblp:conf/naacl/RastogiDA15 fatcat:hfhf265rzne7ndli6vvmwklr7m

Biconditional Generative Adversarial Networks for Multiview Learning with Missing Views [chapter]

Anastasiia Doinychko, Massih-Reza Amini
2020 Lecture Notes in Computer Science  
Our proposed approach addresses this problem by jointly learning the missing views and the multiview classifier using a tripartite game with two generators and a discriminator.  ...  In this paper, we present a conditional GAN with two generators and a common discriminator for multiview learning problems where observations have two views, but one of them may be missing for some of  ...  The former approaches rely mostly on an encoder-encoder network to first map images into a latent space and then generate their views using an inverse mapping.  ... 
doi:10.1007/978-3-030-45439-5_53 fatcat:nlz3s2tmbfcfrdnuuiuqyyqdfe

Large-Margin Predictive Latent Subspace Learning for Multiview Data Analysis

Ning Chen, Jun Zhu, Fuchun Sun, E. P. Xing
2012 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Our approach is based on a multiview latent subspace Markov network (MN) which fulfills a weak conditional independence assumption that multiview observations and response variables are conditionally independent  ...  given a set of latent variables.  ...  Part of this work was done while Ning Chen was a  ... 
doi:10.1109/tpami.2012.64 pmid:22392706 fatcat:4vizascr4jddxjnr443sncjn7e

Adversarial Canonical Correlation Analysis [article]

Benjamin Dutton
2020 arXiv   pre-print
Canonical Correlation Analysis (CCA) is a statistical technique used to extract common information from multiple data sources or views.  ...  We offer further analysis on the multi-level disentangling properties of VCCA-Private and ACCA-Private through the use of a newly designed dataset we call Tangled MNIST.  ...  ( a ) a View x generations in VCCA (b) View y generations in VCCA (c) View x generations in ACCA (d) View y generations in ACCA Fig. 11: Here we explore the information content in z as a function of  ... 
arXiv:2005.10349v2 fatcat:lr2rovbgszfgpm6ids42i3gmca

On Multiview Analysis for Fingerprint Liveness Detection [chapter]

Amirhosein Toosi, Sandro Cumani, Andrea Bottino
2015 Lecture Notes in Computer Science  
In this work, we present the results of a preliminary investigation on multiview analysis for fingerprint liveness detection.  ...  Several approaches to discriminate between live and fake fingerprint images have been presented to address this issue.  ...  Static methods, on the contrary, focus their analysis on a single fingerprint image, which makes them more general and attractive. These methods can be, again, divided into two main categories.  ... 
doi:10.1007/978-3-319-25751-8_18 fatcat:q7eq75os2vfmlgmngxlxe2johi

Variational Inference for Deep Probabilistic Canonical Correlation Analysis [article]

Mahdi Karami, Dale Schuurmans
2020 arXiv   pre-print
In this paper, we propose a deep probabilistic multi-view model that is composed of a linear multi-view layer based on probabilistic canonical correlation analysis (CCA) description in the latent space  ...  A generalization to models with arbitrary number of views is also proposed.  ...  A deep generative probabilistic model for multiview data was studied.  ... 
arXiv:2003.04292v1 fatcat:hzxjrysrebacbaivca5rpp335q

Social Media Meets Big Urban Data: A Case Study of Urban Waterlogging Analysis

Ningyu Zhang, Huajun Chen, Jiaoyan Chen, Xi Chen
2016 Computational Intelligence and Neuroscience  
Moreover, we use a multiview discriminant transfer learning method to transfer knowledge to small cities.  ...  In this paper, we propose a transfer learning method for urban waterlogging disaster analysis, which provides the basis for traffic management agencies to generate proactive traffic operation strategies  ...  [13] , based on geotagged social media, the authors proposed a multilevel generative model that reasons jointly about latent topics and geographical regions.  ... 
doi:10.1155/2016/3264587 pmid:27774098 pmcid:PMC5059775 fatcat:coam3g4ucfdtvcnv25ipkg2qpa

Deep Generalized Canonical Correlation Analysis [article]

Adrian Benton, Huda Khayrallah, Biman Gujral, Dee Ann Reisinger, Sheng Zhang, Raman Arora
2017 arXiv   pre-print
We present Deep Generalized Canonical Correlation Analysis (DGCCA) -- a method for learning nonlinear transformations of arbitrarily many views of data, such that the resulting transformations are maximally  ...  multiview representation learning technique that combines the flexibility of nonlinear (deep) representation learning with the statistical power of incorporating information from many independent sources  ...  Here we present Deep Generalized Canonical Correlation Analysis (DGCCA).  ... 
arXiv:1702.02519v2 fatcat:3omuyay7nbadzibh3jtwz6yfhy

Semi-paired Probabilistic Canonical Correlation Analysis [chapter]

Bo Zhang, Jie Hao, Gang Ma, Jinpeng Yue, Zhongzhi Shi
2014 IFIP Advances in Information and Communication Technology  
Experiments with artificially generated samples demonstrate the effectiveness of the proposed method.  ...  CCA is a powerful tool for analyzing paired multi-view data.  ...  As a discriminant function, we used f x a T x θ where a a , … , a T , and θ is the discrimination threshold such that the larger θ we set, the more samples removed.  ... 
doi:10.1007/978-3-662-44980-6_1 fatcat:lkmcaimklzhubi2i62qzjd64ky

Unsupervised Correlation Analysis

Yedid Hoshen, Lior Wolf
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We introduce a new method Unsupervised Correlation Analysis (UCA), which requires no prior correspondences between the two domains.  ...  One of the most popular methods for linking between domains is Canonical Correlation Analysis (CCA). All current CCA algorithms require correspondences between the views.  ...  It does not use a shared latent space, and the shared space effectively becomes the original target space.  ... 
doi:10.1109/cvpr.2018.00350 dblp:conf/cvpr/HoshenW18 fatcat:hk6lccjsprggdnfgstffb2ngma

Unsupervised Correlation Analysis [article]

Yedid Hoshen, Lior Wolf
2018 arXiv   pre-print
We introduce a new method Unsupervised Correlation Analysis (UCA), which requires no prior correspondences between the two domains.  ...  One of the most popular methods for linking between domains is Canonical Correlation Analysis (CCA). All current CCA algorithms require correspondences between the views.  ...  It does not use a shared latent space, and the shared space effectively becomes the original target space.  ... 
arXiv:1804.00347v1 fatcat:7yv6jklvrfaxtoinb5kdnb6oou

Multi-view Common Component Discriminant Analysis for Cross-view Classification [article]

Xinge You, Jiamiao Xu, Wei Yuan, Xiao-Yuan Jing, Dacheng Tao and Taiping Zhang
2018 arXiv   pre-print
To circumvent this drawback, we propose Multi-view Common Component Discriminant Analysis (MvCCDA) to handle view discrepancy, discriminability and nonlinearity in a joint manner.  ...  We develop a kernel method of MvCCDA to further boost the performance of MvCCDA. Beyond kernel extension, optimization and complexity analysis of MvCCDA are also presented for completeness.  ...  To overcome this shortcoming, Generalized Multiview Analysis (GMA) [3] extends MCCA to a supervised algorithm by taking into consideration intra-view discriminant information.  ... 
arXiv:1805.05029v2 fatcat:53qo6ppr4zh2vfupkafnjxhe2i
« Previous Showing results 1 — 15 out of 1,049 results