Multi-View Learning in the Presence of View Disagreement
arXiv (Cornell University)Published 13 June 2012Open access
C. Mario Christoudias, Raquel Urtasun, Trevor Darrell
Citations96
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
Traditional multi-view learning approaches suffer in the presence of view disagreement,i.e., when samples in each view do not belong to the same class due to view corruption, occlusion or other noise processes. In this paper we present a multi-view learning approach that uses a conditional entropy criterion to detect view disagreement. Once detected, samples with view disagreement are filtered and standard multi-view learning methods can be successfully applied to the remaining samples. Experimental evaluation on synthetic and audio-visual databases demonstrates that the detection and filtering of view disagreement considerably increases the performance of traditional multi-view learning approaches.
Keywords
Computer ScienceEngineering
Elements of Information Theory
37,533 Citations2001Thomas M. Cover, Joy A. Thomas
Density Estimation for Statistics and Data Analysis
15,768 Citations2018B.W. Silverman
Combining labeled and unlabeled data with co-training
5,604 Citations1998Avrim Blum, Tom M. Mitchell
Co-clustering documents and words using bipartite spectral graph partitioning
1,696 Citations2001Inderjit S. Dhillon
A new spectral co-clustering algorithm is used that uses the second left and right singular vectors of an appropriately scaled word-document matrix to yield good bipartitionings and it can be shown that the singular vectors solve a real relaxation to the NP-complete graph bipartitionsing problem.
Analyzing the effectiveness and applicability of co-training
1,028 Citations2000Kamal Nigam, Rayid Ghani
It is demonstrated that when learning from labeled and unlabeled data, algorithms explicitly leveraging a natural independent split of the features outperform algorithms that do not and may out-perform algorithms not using a split.
Information-theoretic co-clustering
998 Citations2003Inderjit S. Dhillon, Subramanyam Mallela +1 more
This work presents an innovative co-clustering algorithm that monotonically increases the preserved mutual information by intertwining both the row and column clusterings at all stages and demonstrates that the algorithm works well in practice, especially in the presence of sparsity and high-dimensionality.
Unsupervised Models for Named Entity Classification
817 Citations1999Michael Collins, Yoram Singer
It is shown that the use of unlabeled data can reduce the requirements for supervision to just 7 simple "seed" rules, gaining leverage from natural redundancy in the data.
A Co-Regularization Approach to Semi-supervised Learning with Multiple Views
337 Citations2005Vikas Sindhwani, Partha Niyogi +1 more
This paper proposes a Co-Regularization framework where classifiers are learnt in each view through forms of multi-view regularization, and proposes algorithms within this framework that are based on optimizing measures of agreement and smoothness over labeled and unlabeled examples.
The MIT Press eBooksPAC Generalization Bounds for Co-training
287 Citations2002Sanjoy Dasgupta, Michael L. Littman +1 more
A new PAC-style bound on generalization error is given which justifies both the use of confidences — partial rules and partial labeling of the unlabeled data — and theUse of an agreement-based objective function as suggested by Collins and Singer.
Sparse probabilistic regression for activity-independent human pose inference
206 Citations2008Raquel Urtasun, Trevor Darrell
An online probabilistic regression scheme for efficient inference of complex, high- dimensional, and multimodal mappings, based on a local mixture of Gaussian processes, where locality is defined based on both appearance and pose, and where the mapping hyperparameters can vary across local neighborhoods to better adapt to specific regions in the pose space.
Lecture notes in computer scienceMulti-view Regression Via Canonical Correlation Analysis
192 Citations2007Sham M. Kakade, Dean P. Foster
This work provides a semi-supervised algorithm which first uses unlabeled data to learn a norm (or, equivalently, a kernel) and then uses labeled data in a ridge regression algorithm (with this induced norm) to provide the predictor.
Adaptive view-based appearance models
107 Citations2003L.-P. Morency, Asgar Bradaran Rahimi +1 more
This work presents a method for online rigid object tracking using an adaptive view-based appearance model that has bounded drift and can track objects undergoing large motion for long periods of time when the object's pose trajectory crosses itself.
Two-view feature generation model for semi-supervised learning
95 Citations2007Rie Kubota Ando, Tong Zhang
The two-view feature generation model of co-training is revisited and it is proved that the optimum predictor can be expressed as a linear combination of a few features constructed from unlabeled data.
Heterogeneous acoustic measurements and multiple classifiers for speech recognition
90 Citations1999Andrew K. Halberstadt, James Glass
Pyrotinib showed activity against NSCLC with HER2 exon 20 mutations in both patient-derived organoids and a PDX model and showed promising efficacy in the clinical trial.
Semi-Supervised Cross Feature Learning for Semantic Concept Detection in Videos
75 Citations2005Rong Yan, Milind Naphade
The proposed SCFL algorithm not only significantly outperforms the conventional co-training algorithms, but also comes close to achieving the performance when the unlabeled set were to be manually annotated and used for training along with the labeled data set.
Bayesian Co-Training
56 Citations2007Shipeng Yu, Balaji Krishnapuram +3 more
Lecture notes in computer scienceEstimation of Mixture Models Using Co-EM
55 Citations2005Steffen Bickel, Tobias Scheffer
This work studies estimation of mixture models for problems in which multiple views of the instances are available, and derives an algorithm that maximizes the likelihood and the consensus among models in the individual views.
Adaptive View Validation: A First Step Towards Automatic View Detection
41 Citations2002Ion Muslea, Steven Minton +1 more
This work uses information acquired while solving several exemplar learning tasks to train a classifier that discriminates between the tasks for which the views are sufficiently and insufficiently compatible for multi-view learning.
Co-Adaptation of audio-visual speech and gesture classifiers
32 Citations2006C. Mario Christoudias, Kate Saenko +2 more
It is demonstrated that multimodal co-training can be used to learn from only a few labeled examples in one or both of the audio- visual modalities, and a co-adaptation algorithm is proposed, which adapts existing audio-visual classifiers to a particular user or noise condition by leveraging the redundancy in the unlabeled data.
Estimating Dependency and Significance for High-Dimensional Data
4 Citations2006Michael R. Siracusa, Kinh Tieu +3 more
The use of nonparametric tests for characterizing dependency and how to carry out these tests with high-dimensional observations are reviewed and a method to assess the significance of the tests is presented.