SlideShare une entreprise Scribd logo
1  sur  35
SemiBoost : Boosting for Semi-supervised Learning Pavan Kumar Mallapragada, Student Member, IEEE, Rong Jin, Member, IEEE, Anil K. Jain, Fellow, IEEE, and Yi Liu, Student Member, IEEE Presented by Yueng-Tien ,Lo Reference: SemiBoost: Boosting for Semi-supervised Learning  P. K. Mallapragada, R. Jin, A. K. Jain, and Y. Liu,  IEEE Transaction on Pattern Analysis and Machine Intelligence(PAMI), 31(11):2000-2014, 2009
outline Introduction Related Work Semi-Supervised Boosting Results and Discussion Conclusions and Future Work Gaussian Fields and Harmonic Functions
Introduction The key idea of semi-supervised learning, specifically semi-supervised classification, is to exploit both labeled and unlabeled data to learn a classification model. There is an immense need for algorithms that can utilize the small amount of labeled data, combined with the large amount of unlabeled data to build efficient classification systems.
Introduction Existing semi-supervised classification algorithms may be classified into two categories based on their underlying assumptions. manifold assumption  cluster assumption
Introduction manifold assumption : the data lie on a low dimensional manifold in the input space cluster assumption : the data samples with high similarity between them must share the same label.
Introduction Most semi-supervised learning approaches design specialized learning algorithms to effectively utilize both labeled and unlabeled data. We refer to this problem of improving the performance of any supervised learning algorithm using unlabeled data as Semi-supervise Improvement, to distinguish our work from the standard semi-supervised learning problems.
Introduction The key difficulties in designing SemiBoost are:  how to sample the unlabeled examples for training a new classification model at each iteration  what class labels should be assigned to the selected unlabeled examples.
Introduction One way to address the above questions is to exploit both the clustering assumption and the large margin criterion Selecting the unlabeled examples with the highest classification confidence and assign them the class labels that are predicted by the current classifier. A problem with this strategy is that the introduction of examples with predicted class labels may only help to increase the classification margin, without actually providing any novel information to the classifier.
Introduction To overcome the above problem, we propose using the pairwise similarity measurements to guide the selection of unlabeled examples at each iteration, as well as for assigning class labels to them.
Related Work
Related Work An inductive algorithm can be used to predict the labels of samples that are unseen during training (irrespective of it being labeled or unlabeled). Transductivealgorithms are limited to predicting only the labels of the unlabeled samples seen during training.
Related Work A popular way to define the inconsistency between the labels                of the samples           and the pairwise similarities Sij is the quadratic criterion: 	where L is the combinatorial graph Laplacian. The task is to assign values to the unknown labels in such a way that the overall inconsistency is minimized.
Semi-Supervised Improvement denote the entire data set, including both the labeled and the unlabeled examples. the first nl examples are labeled, given by  the imputed class labels of unlabeled examples Let                 denote the symmetric similarity matrix, where              represents the similarity between xi and xj.        denote  the              submatrix of the similarity matrix        denote the               submatrix of the similarity matrix ‘A’ denote the given supervised learning algorithm
Semi-Supervised Improvement The goal of semi-supervised improvement is to improve the performance of A iteratively by treating A like a black box, using the unlabeled examples and the pairwise similarity S. In the semi-supervised improvement problem, we aim to build an ensemble classifier which utilizes the unlabeled samples in the way a graph-based approach would utilize.
An outline of the SemiBoost algorithm for semi-supervised improvement Start with an empty ensemble. At each iteration, Compute the peusdo label (and its confidence) for each unlabeled example (using existing ensemble, 	and the pairwise similarity). Sample most confident pseudo labeled examples; combine them with the labeled samples and 	train a component classifier using the supervised  	learning algorithm A. Update the ensemble by including the component classifier with an appropriate weight.
SemiBoost The unlabeled samples must be assigned labels following two main criteria:  The points with high similarity among unlabeled samples must share the same label   Those unlabeled samples which are highly similar to a labeled sample must share its label. Our objective function            is a combination of two terms:  one measuring the inconsistency between labeled and unlabeled examples                and the other measuring the inconsistency among the unlabeled examples
SemiBoost Inspired by the harmonic function approach, we define               , the inconsistency between class labels y and the similarity measurement S, as Note that (1) can be expanded as 	and due to the symmetry of S
SemiBoost We have 	where                                                                     is the hyperbolic cosine function. Rewriting (1) using the               function reveals the connection between the quadratic penalty used in the graph-Laplacian-based approaches, and the exponential penalty used in the current approach
SemiBoost Using a               penalty not only facilitates the derivation of boosting-based algorithms but also increases the classification margin. The inconsistency between labeled and unlabeled examples              is defined as
SemiBoost Combining (1) and (3) leads to the objective function, The constant C is introduced to weight the importance between the labeled and the unlabeled data.  Given the objective function in (4), the optimal class label yu is found by minimizing F.
SemiBoost The problem can now be formally expressed as The following procedure is adopted to derive the boosting algorithm: The labels for the unlabeled samples yiuare replaced by the ensemble predictions over the corresponding data sample. A bound-optimization-based approach is then used to find the ensemble classifier minimizing the objective function. The bounds are simplified further to obtain the sampling scheme, and other required parameters.
The SemiBoost algorithm Compute the pairwise similarity Si,j between any two examples. Initialize H(x) = 0 For  t = 1, 2, . . . , T Compute pi and qi for every example using Equations (9)and(10) Compute the class label zi= sign(pi − qi) for each example Sample example xi by the weight |pi − qi| Apply the algorithm A to train a binary classifier ht(x) using the sampled examples and their class labels zi Compute αt using Equation (11) Update the classification function as H(x) ← H(x) + αtht(x)
SemiBoost-Algorithm Let                                denote the two-class classifi-cation model that is learned at the t-th iteration by the algorithm A.  Let                        denote the combined classification model learned after the first T iterations. 	where        is the combination weight.
SemiBoost-Algorithm ,[object Object]
This expression involves products of variables     and      making it nonlinear and, hence , difficult to optimize.,[object Object]
SemiBoost-Algorithm (prop.1) Minimizing (7) is equivalent to minimizing the function 	where 	and
Appendix
SemiBoost-Algorithm (prop.2) The expression in (8) is difficult to optimize since the weight  and the classifier h(x) are coupled together. Minimizing (8) is equivalent to minimizing We denote the upper bound in the above equation by F2.
SemiBoost-Algorithm (prop.3) To minimize F2, the optimal class label zi for the example xi is zi= sign(pi – qi) and the weight for sampling example xi is |pi-qi|. The optimal       that minimizes F1 is
SemiBoost-Algorithm At each relaxation, the “touch-point” is maintained between the objective function and the upper bound.  As a result, the procedure guarantees:  The objective function always decreases through iterations The final solution converges to a local minimum.
The SemiBoost algorithm Compute the pairwise similarity Si,j between any two examples. Initialize H(x) = 0 For  t = 1, 2, . . . , T Compute pi and qi for every example using Equations (9)and(10) Compute the class label zi= sign(pi − qi) for each example Sample example xi by the weight |pi − qi| Apply the algorithm A to train a binary classifier ht(x) using the sampled examples and their class labels zi Compute αt using Equation (11) Update the classification function as H(x) ← H(x) + αtht(x)
SemiBoost-Algorithm Let       be the weighted error made by the classifier, where As in the case of AdaBoost,     can be expressed as  	which is very similar to the weighting factor of AdaBoost, differing only by a constant factor of 1/2 .
SemiBoost-Algorithm Theorem 1. Let                  be the combination weights that are computed by running the SemiBoost algorithm (Fig. 1). Then, the objective function at the et t 1Tst iteration, i.e., Ftt1, is bounded as follows: where
Results and Discussion
X. Zhu, Z. Ghahramani, and J. Lafferty, “Semi-supervised learning using gaussian fields and harmonic functions,” in Proc. 20t International Conference on Machine Learning, pp. 912–919, 2003

Contenu connexe

Tendances

Inference & Learning in Linear Chain Conditional Random Fields (CRFs)
Inference & Learning in Linear Chain Conditional Random Fields (CRFs)Inference & Learning in Linear Chain Conditional Random Fields (CRFs)
Inference & Learning in Linear Chain Conditional Random Fields (CRFs)Anmol Dwivedi
 
Machine Learning Unit 3 Semester 3 MSc IT Part 2 Mumbai University
Machine Learning Unit 3 Semester 3  MSc IT Part 2 Mumbai UniversityMachine Learning Unit 3 Semester 3  MSc IT Part 2 Mumbai University
Machine Learning Unit 3 Semester 3 MSc IT Part 2 Mumbai UniversityMadhav Mishra
 
Master of Computer Application (MCA) – Semester 4 MC0079
Master of Computer Application (MCA) – Semester 4  MC0079Master of Computer Application (MCA) – Semester 4  MC0079
Master of Computer Application (MCA) – Semester 4 MC0079Aravind NC
 
Instance based learning
Instance based learningInstance based learning
Instance based learningswapnac12
 
Mc0079 computer based optimization methods--phpapp02
Mc0079 computer based optimization methods--phpapp02Mc0079 computer based optimization methods--phpapp02
Mc0079 computer based optimization methods--phpapp02Rabby Bhatt
 
Tutorial on Markov Random Fields (MRFs) for Computer Vision Applications
Tutorial on Markov Random Fields (MRFs) for Computer Vision ApplicationsTutorial on Markov Random Fields (MRFs) for Computer Vision Applications
Tutorial on Markov Random Fields (MRFs) for Computer Vision ApplicationsAnmol Dwivedi
 
SVD and the Netflix Dataset
SVD and the Netflix DatasetSVD and the Netflix Dataset
SVD and the Netflix DatasetBen Mabey
 
lecture_mooney.ppt
lecture_mooney.pptlecture_mooney.ppt
lecture_mooney.pptbutest
 
Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...
Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...
Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...orajjournal
 
110726IGARSS_MIL.pptx
110726IGARSS_MIL.pptx110726IGARSS_MIL.pptx
110726IGARSS_MIL.pptxgrssieee
 
Similarity Features, and their Role in Concept Alignment Learning
Similarity Features, and their Role in Concept Alignment Learning Similarity Features, and their Role in Concept Alignment Learning
Similarity Features, and their Role in Concept Alignment Learning Shenghui Wang
 
Chapter 3.0 new
Chapter 3.0 newChapter 3.0 new
Chapter 3.0 newsotlsoc
 
Machine learning and Neural Networks
Machine learning and Neural NetworksMachine learning and Neural Networks
Machine learning and Neural Networksbutest
 
FOCUS.doc
FOCUS.docFOCUS.doc
FOCUS.docbutest
 

Tendances (15)

Inference & Learning in Linear Chain Conditional Random Fields (CRFs)
Inference & Learning in Linear Chain Conditional Random Fields (CRFs)Inference & Learning in Linear Chain Conditional Random Fields (CRFs)
Inference & Learning in Linear Chain Conditional Random Fields (CRFs)
 
Machine Learning Unit 3 Semester 3 MSc IT Part 2 Mumbai University
Machine Learning Unit 3 Semester 3  MSc IT Part 2 Mumbai UniversityMachine Learning Unit 3 Semester 3  MSc IT Part 2 Mumbai University
Machine Learning Unit 3 Semester 3 MSc IT Part 2 Mumbai University
 
Master of Computer Application (MCA) – Semester 4 MC0079
Master of Computer Application (MCA) – Semester 4  MC0079Master of Computer Application (MCA) – Semester 4  MC0079
Master of Computer Application (MCA) – Semester 4 MC0079
 
Instance based learning
Instance based learningInstance based learning
Instance based learning
 
Mc0079 computer based optimization methods--phpapp02
Mc0079 computer based optimization methods--phpapp02Mc0079 computer based optimization methods--phpapp02
Mc0079 computer based optimization methods--phpapp02
 
Tutorial on Markov Random Fields (MRFs) for Computer Vision Applications
Tutorial on Markov Random Fields (MRFs) for Computer Vision ApplicationsTutorial on Markov Random Fields (MRFs) for Computer Vision Applications
Tutorial on Markov Random Fields (MRFs) for Computer Vision Applications
 
SVD and the Netflix Dataset
SVD and the Netflix DatasetSVD and the Netflix Dataset
SVD and the Netflix Dataset
 
lecture_mooney.ppt
lecture_mooney.pptlecture_mooney.ppt
lecture_mooney.ppt
 
Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...
Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...
Solving Linear Fractional Programming Problems Using a New Homotopy Perturbat...
 
Machine learning
Machine learningMachine learning
Machine learning
 
110726IGARSS_MIL.pptx
110726IGARSS_MIL.pptx110726IGARSS_MIL.pptx
110726IGARSS_MIL.pptx
 
Similarity Features, and their Role in Concept Alignment Learning
Similarity Features, and their Role in Concept Alignment Learning Similarity Features, and their Role in Concept Alignment Learning
Similarity Features, and their Role in Concept Alignment Learning
 
Chapter 3.0 new
Chapter 3.0 newChapter 3.0 new
Chapter 3.0 new
 
Machine learning and Neural Networks
Machine learning and Neural NetworksMachine learning and Neural Networks
Machine learning and Neural Networks
 
FOCUS.doc
FOCUS.docFOCUS.doc
FOCUS.doc
 

En vedette

kantorNSF-NIJ-ISI-03-06-04.ppt
kantorNSF-NIJ-ISI-03-06-04.pptkantorNSF-NIJ-ISI-03-06-04.ppt
kantorNSF-NIJ-ISI-03-06-04.pptbutest
 
Machine Learning and Statistical Analysis
Machine Learning and Statistical AnalysisMachine Learning and Statistical Analysis
Machine Learning and Statistical Analysisbutest
 
Preview Class Handout "
Preview Class Handout "Preview Class Handout "
Preview Class Handout "butest
 
Best Weight Loss Plan For 2010
Best Weight Loss Plan For 2010Best Weight Loss Plan For 2010
Best Weight Loss Plan For 2010Brians Garage Sale
 
Artem Volftrub анатомия интернет банка
Artem Volftrub анатомия интернет банкаArtem Volftrub анатомия интернет банка
Artem Volftrub анатомия интернет банкаguest092df8
 
osm.cs.byu.edu
osm.cs.byu.eduosm.cs.byu.edu
osm.cs.byu.edubutest
 

En vedette (7)

kantorNSF-NIJ-ISI-03-06-04.ppt
kantorNSF-NIJ-ISI-03-06-04.pptkantorNSF-NIJ-ISI-03-06-04.ppt
kantorNSF-NIJ-ISI-03-06-04.ppt
 
Machine Learning and Statistical Analysis
Machine Learning and Statistical AnalysisMachine Learning and Statistical Analysis
Machine Learning and Statistical Analysis
 
Preview Class Handout "
Preview Class Handout "Preview Class Handout "
Preview Class Handout "
 
Best Weight Loss Plan For 2010
Best Weight Loss Plan For 2010Best Weight Loss Plan For 2010
Best Weight Loss Plan For 2010
 
Artem Volftrub анатомия интернет банка
Artem Volftrub анатомия интернет банкаArtem Volftrub анатомия интернет банка
Artem Volftrub анатомия интернет банка
 
osm.cs.byu.edu
osm.cs.byu.eduosm.cs.byu.edu
osm.cs.byu.edu
 
New Media Language
New Media LanguageNew Media Language
New Media Language
 

Similaire à SemiBoost: Boosting for Semi-supervised Learning

Figure 1.doc
Figure 1.docFigure 1.doc
Figure 1.docbutest
 
Figure 1.doc
Figure 1.docFigure 1.doc
Figure 1.docbutest
 
Figure 1.doc
Figure 1.docFigure 1.doc
Figure 1.docbutest
 
activelearning.ppt
activelearning.pptactivelearning.ppt
activelearning.pptbutest
 
Learning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification DataLearning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification Data萍華 楊
 
CSE545_Porject
CSE545_PorjectCSE545_Porject
CSE545_Porjecthan li
 
A Modified KS-test for Feature Selection
A Modified KS-test for Feature SelectionA Modified KS-test for Feature Selection
A Modified KS-test for Feature SelectionIOSR Journals
 
CFM Challenge - Course Project
CFM Challenge - Course ProjectCFM Challenge - Course Project
CFM Challenge - Course ProjectKhalilBergaoui
 
Supervised Learning.pdf
Supervised Learning.pdfSupervised Learning.pdf
Supervised Learning.pdfgadissaassefa
 
Adapted Branch-and-Bound Algorithm Using SVM With Model Selection
Adapted Branch-and-Bound Algorithm Using SVM With Model SelectionAdapted Branch-and-Bound Algorithm Using SVM With Model Selection
Adapted Branch-and-Bound Algorithm Using SVM With Model SelectionIJECEIAES
 
Ai_Project_report
Ai_Project_reportAi_Project_report
Ai_Project_reportRavi Gupta
 
Chapter3 hundred page machine learning
Chapter3 hundred page machine learningChapter3 hundred page machine learning
Chapter3 hundred page machine learningmustafa sarac
 
IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...
IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...
IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...IRJET Journal
 
Lawry-Daniel.doc
Lawry-Daniel.docLawry-Daniel.doc
Lawry-Daniel.docbutest
 
Data.Mining.C.6(II).classification and prediction
Data.Mining.C.6(II).classification and predictionData.Mining.C.6(II).classification and prediction
Data.Mining.C.6(II).classification and predictionMargaret Wang
 
Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm
	Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm	Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm
Study on Evaluation of Venture Capital Based onInteractive Projection Algorithminventionjournals
 
INTRODUCTION TO BOOSTING.ppt
INTRODUCTION TO BOOSTING.pptINTRODUCTION TO BOOSTING.ppt
INTRODUCTION TO BOOSTING.pptBharatDaiyaBharat
 
result analysis for deep leakage from gradients
result analysis for deep leakage from gradientsresult analysis for deep leakage from gradients
result analysis for deep leakage from gradients國騰 丁
 

Similaire à SemiBoost: Boosting for Semi-supervised Learning (20)

Figure 1.doc
Figure 1.docFigure 1.doc
Figure 1.doc
 
Figure 1.doc
Figure 1.docFigure 1.doc
Figure 1.doc
 
Figure 1.doc
Figure 1.docFigure 1.doc
Figure 1.doc
 
activelearning.ppt
activelearning.pptactivelearning.ppt
activelearning.ppt
 
Learning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification DataLearning On The Border:Active Learning in Imbalanced classification Data
Learning On The Border:Active Learning in Imbalanced classification Data
 
CSE545_Porject
CSE545_PorjectCSE545_Porject
CSE545_Porject
 
A Modified KS-test for Feature Selection
A Modified KS-test for Feature SelectionA Modified KS-test for Feature Selection
A Modified KS-test for Feature Selection
 
CFM Challenge - Course Project
CFM Challenge - Course ProjectCFM Challenge - Course Project
CFM Challenge - Course Project
 
Supervised Learning.pdf
Supervised Learning.pdfSupervised Learning.pdf
Supervised Learning.pdf
 
18.1 combining models
18.1 combining models18.1 combining models
18.1 combining models
 
Adapted Branch-and-Bound Algorithm Using SVM With Model Selection
Adapted Branch-and-Bound Algorithm Using SVM With Model SelectionAdapted Branch-and-Bound Algorithm Using SVM With Model Selection
Adapted Branch-and-Bound Algorithm Using SVM With Model Selection
 
Ai_Project_report
Ai_Project_reportAi_Project_report
Ai_Project_report
 
Chapter3 hundred page machine learning
Chapter3 hundred page machine learningChapter3 hundred page machine learning
Chapter3 hundred page machine learning
 
IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...
IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...
IRJET- Evaluation of Classification Algorithms with Solutions to Class Imbala...
 
Lawry-Daniel.doc
Lawry-Daniel.docLawry-Daniel.doc
Lawry-Daniel.doc
 
Data.Mining.C.6(II).classification and prediction
Data.Mining.C.6(II).classification and predictionData.Mining.C.6(II).classification and prediction
Data.Mining.C.6(II).classification and prediction
 
Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm
	Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm	Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm
Study on Evaluation of Venture Capital Based onInteractive Projection Algorithm
 
Building the Professional of 2020: An Approach to Business Change Process Int...
Building the Professional of 2020: An Approach to Business Change Process Int...Building the Professional of 2020: An Approach to Business Change Process Int...
Building the Professional of 2020: An Approach to Business Change Process Int...
 
INTRODUCTION TO BOOSTING.ppt
INTRODUCTION TO BOOSTING.pptINTRODUCTION TO BOOSTING.ppt
INTRODUCTION TO BOOSTING.ppt
 
result analysis for deep leakage from gradients
result analysis for deep leakage from gradientsresult analysis for deep leakage from gradients
result analysis for deep leakage from gradients
 

Plus de butest

EL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEEL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEbutest
 
1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同butest
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALbutest
 
Timeline: The Life of Michael Jackson
Timeline: The Life of Michael JacksonTimeline: The Life of Michael Jackson
Timeline: The Life of Michael Jacksonbutest
 
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...butest
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALbutest
 
Com 380, Summer II
Com 380, Summer IICom 380, Summer II
Com 380, Summer IIbutest
 
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet JazzThe MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazzbutest
 
MICHAEL JACKSON.doc
MICHAEL JACKSON.docMICHAEL JACKSON.doc
MICHAEL JACKSON.docbutest
 
Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1butest
 
Facebook
Facebook Facebook
Facebook butest
 
Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...butest
 
Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...butest
 
NEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTNEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTbutest
 
C-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docC-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docbutest
 
MAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docMAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docbutest
 
Mac OS X Guide.doc
Mac OS X Guide.docMac OS X Guide.doc
Mac OS X Guide.docbutest
 
WEB DESIGN!
WEB DESIGN!WEB DESIGN!
WEB DESIGN!butest
 

Plus de butest (20)

EL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEEL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBE
 
1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
 
Timeline: The Life of Michael Jackson
Timeline: The Life of Michael JacksonTimeline: The Life of Michael Jackson
Timeline: The Life of Michael Jackson
 
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
 
Com 380, Summer II
Com 380, Summer IICom 380, Summer II
Com 380, Summer II
 
PPT
PPTPPT
PPT
 
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet JazzThe MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
 
MICHAEL JACKSON.doc
MICHAEL JACKSON.docMICHAEL JACKSON.doc
MICHAEL JACKSON.doc
 
Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1
 
Facebook
Facebook Facebook
Facebook
 
Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...
 
Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...
 
NEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTNEWS ANNOUNCEMENT
NEWS ANNOUNCEMENT
 
C-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docC-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.doc
 
MAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docMAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.doc
 
Mac OS X Guide.doc
Mac OS X Guide.docMac OS X Guide.doc
Mac OS X Guide.doc
 
hier
hierhier
hier
 
WEB DESIGN!
WEB DESIGN!WEB DESIGN!
WEB DESIGN!
 

SemiBoost: Boosting for Semi-supervised Learning

  • 1. SemiBoost : Boosting for Semi-supervised Learning Pavan Kumar Mallapragada, Student Member, IEEE, Rong Jin, Member, IEEE, Anil K. Jain, Fellow, IEEE, and Yi Liu, Student Member, IEEE Presented by Yueng-Tien ,Lo Reference: SemiBoost: Boosting for Semi-supervised Learning P. K. Mallapragada, R. Jin, A. K. Jain, and Y. Liu, IEEE Transaction on Pattern Analysis and Machine Intelligence(PAMI), 31(11):2000-2014, 2009
  • 2. outline Introduction Related Work Semi-Supervised Boosting Results and Discussion Conclusions and Future Work Gaussian Fields and Harmonic Functions
  • 3. Introduction The key idea of semi-supervised learning, specifically semi-supervised classification, is to exploit both labeled and unlabeled data to learn a classification model. There is an immense need for algorithms that can utilize the small amount of labeled data, combined with the large amount of unlabeled data to build efficient classification systems.
  • 4. Introduction Existing semi-supervised classification algorithms may be classified into two categories based on their underlying assumptions. manifold assumption cluster assumption
  • 5. Introduction manifold assumption : the data lie on a low dimensional manifold in the input space cluster assumption : the data samples with high similarity between them must share the same label.
  • 6. Introduction Most semi-supervised learning approaches design specialized learning algorithms to effectively utilize both labeled and unlabeled data. We refer to this problem of improving the performance of any supervised learning algorithm using unlabeled data as Semi-supervise Improvement, to distinguish our work from the standard semi-supervised learning problems.
  • 7. Introduction The key difficulties in designing SemiBoost are: how to sample the unlabeled examples for training a new classification model at each iteration what class labels should be assigned to the selected unlabeled examples.
  • 8. Introduction One way to address the above questions is to exploit both the clustering assumption and the large margin criterion Selecting the unlabeled examples with the highest classification confidence and assign them the class labels that are predicted by the current classifier. A problem with this strategy is that the introduction of examples with predicted class labels may only help to increase the classification margin, without actually providing any novel information to the classifier.
  • 9. Introduction To overcome the above problem, we propose using the pairwise similarity measurements to guide the selection of unlabeled examples at each iteration, as well as for assigning class labels to them.
  • 11. Related Work An inductive algorithm can be used to predict the labels of samples that are unseen during training (irrespective of it being labeled or unlabeled). Transductivealgorithms are limited to predicting only the labels of the unlabeled samples seen during training.
  • 12. Related Work A popular way to define the inconsistency between the labels of the samples and the pairwise similarities Sij is the quadratic criterion: where L is the combinatorial graph Laplacian. The task is to assign values to the unknown labels in such a way that the overall inconsistency is minimized.
  • 13. Semi-Supervised Improvement denote the entire data set, including both the labeled and the unlabeled examples. the first nl examples are labeled, given by the imputed class labels of unlabeled examples Let denote the symmetric similarity matrix, where represents the similarity between xi and xj. denote the submatrix of the similarity matrix denote the submatrix of the similarity matrix ‘A’ denote the given supervised learning algorithm
  • 14. Semi-Supervised Improvement The goal of semi-supervised improvement is to improve the performance of A iteratively by treating A like a black box, using the unlabeled examples and the pairwise similarity S. In the semi-supervised improvement problem, we aim to build an ensemble classifier which utilizes the unlabeled samples in the way a graph-based approach would utilize.
  • 15. An outline of the SemiBoost algorithm for semi-supervised improvement Start with an empty ensemble. At each iteration, Compute the peusdo label (and its confidence) for each unlabeled example (using existing ensemble, and the pairwise similarity). Sample most confident pseudo labeled examples; combine them with the labeled samples and train a component classifier using the supervised learning algorithm A. Update the ensemble by including the component classifier with an appropriate weight.
  • 16. SemiBoost The unlabeled samples must be assigned labels following two main criteria: The points with high similarity among unlabeled samples must share the same label Those unlabeled samples which are highly similar to a labeled sample must share its label. Our objective function is a combination of two terms: one measuring the inconsistency between labeled and unlabeled examples and the other measuring the inconsistency among the unlabeled examples
  • 17. SemiBoost Inspired by the harmonic function approach, we define , the inconsistency between class labels y and the similarity measurement S, as Note that (1) can be expanded as and due to the symmetry of S
  • 18. SemiBoost We have where is the hyperbolic cosine function. Rewriting (1) using the function reveals the connection between the quadratic penalty used in the graph-Laplacian-based approaches, and the exponential penalty used in the current approach
  • 19. SemiBoost Using a penalty not only facilitates the derivation of boosting-based algorithms but also increases the classification margin. The inconsistency between labeled and unlabeled examples is defined as
  • 20. SemiBoost Combining (1) and (3) leads to the objective function, The constant C is introduced to weight the importance between the labeled and the unlabeled data. Given the objective function in (4), the optimal class label yu is found by minimizing F.
  • 21. SemiBoost The problem can now be formally expressed as The following procedure is adopted to derive the boosting algorithm: The labels for the unlabeled samples yiuare replaced by the ensemble predictions over the corresponding data sample. A bound-optimization-based approach is then used to find the ensemble classifier minimizing the objective function. The bounds are simplified further to obtain the sampling scheme, and other required parameters.
  • 22. The SemiBoost algorithm Compute the pairwise similarity Si,j between any two examples. Initialize H(x) = 0 For t = 1, 2, . . . , T Compute pi and qi for every example using Equations (9)and(10) Compute the class label zi= sign(pi − qi) for each example Sample example xi by the weight |pi − qi| Apply the algorithm A to train a binary classifier ht(x) using the sampled examples and their class labels zi Compute αt using Equation (11) Update the classification function as H(x) ← H(x) + αtht(x)
  • 23. SemiBoost-Algorithm Let denote the two-class classifi-cation model that is learned at the t-th iteration by the algorithm A. Let denote the combined classification model learned after the first T iterations. where is the combination weight.
  • 24.
  • 25.
  • 26. SemiBoost-Algorithm (prop.1) Minimizing (7) is equivalent to minimizing the function where and
  • 28. SemiBoost-Algorithm (prop.2) The expression in (8) is difficult to optimize since the weight and the classifier h(x) are coupled together. Minimizing (8) is equivalent to minimizing We denote the upper bound in the above equation by F2.
  • 29. SemiBoost-Algorithm (prop.3) To minimize F2, the optimal class label zi for the example xi is zi= sign(pi – qi) and the weight for sampling example xi is |pi-qi|. The optimal that minimizes F1 is
  • 30. SemiBoost-Algorithm At each relaxation, the “touch-point” is maintained between the objective function and the upper bound. As a result, the procedure guarantees: The objective function always decreases through iterations The final solution converges to a local minimum.
  • 31. The SemiBoost algorithm Compute the pairwise similarity Si,j between any two examples. Initialize H(x) = 0 For t = 1, 2, . . . , T Compute pi and qi for every example using Equations (9)and(10) Compute the class label zi= sign(pi − qi) for each example Sample example xi by the weight |pi − qi| Apply the algorithm A to train a binary classifier ht(x) using the sampled examples and their class labels zi Compute αt using Equation (11) Update the classification function as H(x) ← H(x) + αtht(x)
  • 32. SemiBoost-Algorithm Let be the weighted error made by the classifier, where As in the case of AdaBoost, can be expressed as which is very similar to the weighting factor of AdaBoost, differing only by a constant factor of 1/2 .
  • 33. SemiBoost-Algorithm Theorem 1. Let be the combination weights that are computed by running the SemiBoost algorithm (Fig. 1). Then, the objective function at the et t 1Tst iteration, i.e., Ftt1, is bounded as follows: where
  • 35. X. Zhu, Z. Ghahramani, and J. Lafferty, “Semi-supervised learning using gaussian fields and harmonic functions,” in Proc. 20t International Conference on Machine Learning, pp. 912–919, 2003