SlideShare a Scribd company logo
1 of 40
Download to read offline
CRF
          2010   12   10       10


      1                    (        )
‣

‣


    2
4
[Lafferty+, 01] Conditional Random Fields: Probabilistic Models
for Segmenting and Labeling Sequence Data. John Lafferty, Andrew
McCallum, Fernando Pereira. Proceedings of ICML’01, 2001.

[Collins, 02] Discriminative training methods for hidden markov
models: Theory and experiments with perceptron algorithms.
Michael Collins. Proceedings of EMNLP’02, 2002.

[Morency+, 07] Latent-dynamic discriminative models for
continuous gesture recognition. Louis-Philippe Morency, Ariadna
Quattoni, and Trevor Darrell. Proceedings of CVPR’07, 2007.

[Sun+, 09] Latent Variable Perceptron Algorithm for Structured
Classification. Xu Sun, Takuya Matsuzaki, Daisuke Okanohara and
Jun’ichi Tsujii. Proceedings of IJCAI’09, 2009                     3
‣
‣

    CRF                                     Structured
    (Conditional Random Field)              Perceptron
                                    1   2
                                    3   4
    DPLVM                                   Latent Variable
    (Discriminative Probabilistic
    Latent Variable Model)                  Perceptron

‣                                                             4
x=   x1        x2               xm



y=   y1        y2               ym

          y1 , . . . , ym ∈ Y

                                     5
(   : NP-chunking)


x1   x2        x3       x4      x5
He   is        her    brother    .



B    O          B        I      O
y1   y2        y3       y4     y5
                        Y = {B, I, O}
                                     6
‣
‣

    CRF                                     Structured
    (Conditional Random Field)              Perceptron
                                    1   2
                                    3   4
    DPLVM                                   Latent Variable
    (Discriminative Probabilistic
    Latent Variable Model)                  Perceptron

‣                                                             7
Θ
                          P (y|x, Θ)
P (yi |xi , Θ)
    ∗

            Θ
                            {(xi , yi )}i=1
                                    ∗ d

                                       ..
                                            .


                 ..
                      .                         d
                                                    8
Θ
    P (y|x, Θ)


x
                        ˆ
          y = argmax P (y|x, Θ)
                 ˆ
                 y




                                  9
(x, y)

         →
                            
        f1 (y, x)         Θ1
    
       f2 (y, x)    
                        Θ2   
                               
           .
            .            .
                           .   
n
           .       ·    .    = F (y|x, Θ)
                            
           .
            .
                         .
                           .
                               
           .            .   
        fn (y, x)         Θn
           =




                          =

        f (y, x)          Θ                     10
1
      P (y|x, Θ) = exp F (y|x, Θ)
                 Z                
                  
               Z=   exp F (y |x, Θ)
                            

                     y

                          F (y|x, Θ) = f (y, x) · Θ
1/Z


      argmax P (y|x, Θ) = argmax F (y|x, Θ)
         y                    y                       11
1
      P (y|x, Θ) = exp F (y|x, Θ)
                 Z                
                  
               Z=   exp F (y |x, Θ)
                            

                     y

                          F (y|x, Θ) = fO(|Yx) ) Θ
                                        (y, |m ·
1/Z


      argmax P (y|x, Θ) = argmax F (y|x, Θ)
         y                    y                      12
CRF: Conditional Random Field (sequential)

              yj−1             yj
             
                           s(j, x, yj )
                 t(j, x, yj−1 , yj )
                      ⇒




                                          13
CRF: Conditional Random Field (sequential)

              yj−1             yj
             
                           s(j, x, yj )
                 t(j, x, yj−1 , yj )
                      ⇒




                                          14
CRF


                 d
                 
      maximize         log P (yi |xi , Θ)
                               ∗
                                            − R(Θ)
                 i=1

                 R(Θ)                 Θ




                                                     15
‣
‣

    CRF                                     Structured
    (Conditional Random Field)              Perceptron
                                    1   2
                                    3   4
    DPLVM                                   Latent Variable
    (Discriminative Probabilistic
    Latent Variable Model)                  Perceptron

‣                                                             16
Structured Perceptron

‣
‣       (xi , yi )
               ∗
                               F (yi |xi , Θ)
                                   ∗
                                                =Θ·      f (yi , xi )
                                                             ∗




(xi , yi )
       ∗


                          yi = argmax F (y|xi , Θ )          i
                                        y

               yi =      ∗
                         yi                                      yi =    ∗
                                                                        yi

    Θ   i+1
              =Θ +   i
                         f (yi , xi )
                             ∗
                                        − f (yi , xi )    Θ   i+1
                                                                    =Θ   i

                                                                             17
Structured Perceptron


                    Θ   i+1
                              =Θ +     i
                                            f (yi , xi )
                                                ∗
                                                                − f (yi , xi )



    Θ   i+1
              ·   (f (yi , xi )
                       ∗
                                  − f (yi , xi ))
                                                                                           2
    =Θ ·      i
                   (f (yi , xi )
                        ∗
                                   − f (yi , xi )) +      f (yi , xi )
                                                               ∗
                                                                          −   f (yi , xi )2

⇔   F (yi |xi , Θ )
        ∗        i+1
                              − F (yi |xi , Θ       i+1
                                                          )
                                                                                               2
    =    F (yi |xi , Θi )
             ∗
                                  − F (yi |xi , Θ ) +
                                                    i
                                                              f (yi , xi )
                                                                   ∗
                                                                              −   f (yi , xi )2
                                                                                          ≥0
                                                                                                   18
Structured Perceptron


            Θ   i+1
                      =Θ +  i
                                 f (yi , xi )
                                     ∗
                                                    − f (yi , xi )


                   ∗
                  yi                                                            yi


 F (yi |xi , Θ )
     ∗        i+1
                      − F (yi |xi , Θ   i+1
                                              )
                                                                                   2
 =   F (yi |xi , Θi )
         ∗
                        − F (yi |xi , Θ ) +
                                        i
                                                  f (yi , xi )
                                                       ∗
                                                                  −   f (yi , xi )2
                                                                              ≥0
                                                                                       19
Structured Perceptron

‣
‣




                d
                        M


                            20
separability

G(xi ) = {all possible label sequences for an example xi },
G(xi ) = G(xi ) −     ∗
                    {yi }



        {(xi , yi )}d
                ∗
                    i=1              δ0
     U2 = 1                   U
     ∀i, ∀z ∈ G(xi ), F (yi |xi , U) − F (z|xi , U) ≥ δ.
                          ∗




                                                              21
mistake bound


                      δ0
{(xi , yi )}d
        ∗
            i=1
                                 M
                                         2
                             R
                           M≤ 2
                              δ
        R         ∀i, ∀z ∈ G(xi ), f (yi , xi ) − f (z, xi )2 ≤ R
                                        ∗




                                     d                                22
‣
‣

    CRF                                     Structured
    (Conditional Random Field)              Perceptron
                                    1   2
                                    3   4
    DPLVM                                   Latent Variable
    (Discriminative Probabilistic
    Latent Variable Model)                  Perceptron

‣                                                             23
They   are    her   flowers   .
 B      O     B       I      O


They   gave   her   flowers   .
 B      O     B       B      O

                                 24
They   are    her        flowers   .
 B      O     B            I      O
                    B1
They   gave   her        flowers   .
 B      O     B            B      O
                    B2
                                      25
DPLVM - Discriminative Probabilistic Latent Variable Model



                   Y ={ B , I , O }
            
            
            
            
            
            
            
            
            
                        
                        
                        
                        
                        
                        
                        
                        
                        
                        
        HB = {      B1 , . . . , B|HB | }



                           |HB |

                                                        26
DPLVM - Discriminative Probabilistic Latent Variable Model



    y=        y1          y2                 ym


    h=        h1          h2                  hm

                    ∀j, hj ∈ Hyj
                            def.
                           ⇐⇒ Proj(h) = y
                                                        27
DPLVM

             (x, h)

            →
                                
            f1 (h, x)         Θ1
           f2 (h, x)       Θ2   
                                
               .
                .            .
                               .   
               .       ·    .    = F (h|x, Θ)
                                
               .            .   
               .
                .            .
                               .   
            fn (h, x)         Θn
               =




                              =

            f (h, x)          Θ                 28
DPLVM

                h
                    1
        P (h|x, Θ) = exp F (h|x, Θ)
                   Z                
                    
                 Z=   exp F (h |x, Θ)
                              

                        h

                      F (h|x, Θ) = f (h, x) · Θ
           f (h, x)
  argmax P (h|x, Θ) = argmax F (h|x, Θ)
    h                        h
                                                  29
DPLVM
               ∗
        (xi , yi )                     h
              P (h|x, Θ)


          ∗
         yi                   h
                     
  P (y|x, Θ) =            P (y|h, x, Θ)P (h|x, Θ)
                      h
                          
              =                    P (h|x, Θ)
                     h:Proj(h)=y

                                                    30
DPLVM


              d
              
   maximize         log P (yi |xi , Θ)
                            ∗
                                         − R(Θ)
              i=1

              R(Θ)                 Θ




                                                  31
‣
‣

    CRF                                     Structured
    (Conditional Random Field)              Perceptron
                                    1   2
                                    3   4
    DPLVM                                   Latent Variable
    (Discriminative Probabilistic
    Latent Variable Model)                  Perceptron

‣                                                             32
Latent Variable Perceptron

 (xi , yi )
        ∗
                   hi = argmax F (hi |xi , Θ),
                                  h
                       yi = Proj(hi )

          yi =    ∗
                  yi                                yi =    ∗
                                                           yi

Θ   i+1
          =Θ +i
                   f (hi , xi )
                       ∗
                                  − f (h, xi ) Θ   i+1
                                                         =Θ   i

               ∗
              hi
               ∗
              hi   =       argmax       F (h|xi , Θ )
                                                    i
                                    ∗
                         h:Proj(h)=yi                             33
mistake bound


                  δ0
{(xi , yi )}i=1
        ∗ d


                                         M

                      2T M       2   2
                   M≤
                        δ2
     T                       d
M = max f (y, xi )2 .
       i,y


                                             34
‣
‣

    CRF                                     Structured
    (Conditional Random Field)              Perceptron
                                    1   2
                                    3   4
    DPLVM                                   Latent Variable
    (Discriminative Probabilistic
    Latent Variable Model)                  Perceptron

‣                                                             35
(                     )
‣         X = {a, b}
‣           Y = {A, B}
‣                     HA = {A1 , A2 }, HB = {B1 , B2 }
‣                                     P (hj |hj−1 )
        P (xj |hj )            h     x
‣            y = Proj(h)

‣         {(xi , yi )}i=1
                  ∗ d




‣                                                        36
(                   )
‣                                   p
    from  to      A1              A2           B1      B2
       A1       (1 − p)/3 (1 − p)/3 (1 − p)/3           p
       A2          p            (1 − p)/3 (1 − p)/3 (1 − p)/3
      B1        (1 − p)/3          p      (1 − p)/3 (1 − p)/3
      B2        (1 − p)/3 (1 − p)/3             p    (1 − p)/3

‣                      P (xi = a|hi )
     hi = A1         hi = A2            hi = B1      hi = B2
        0.1              0.7              0.7          0.6       37
(                       )

                           Latent Variable Perceptron              Structured Perceptron

               100

                90
accuracy [%]




                80

                70

                60

                50

                40
                     0.5   0.55    0.6    0.65    0.7       0.75    0.8    0.85    0.9     0.95
                                                        p
                                                                                                  38
‣

‣


‣

‣


‣
    39
‣



‣




    40

More Related Content

What's hot

修士論文発表会
修士論文発表会修士論文発表会
修士論文発表会Keikusl
 
Cosmin Crucean: Perturbative QED on de Sitter Universe.
Cosmin Crucean: Perturbative QED on de Sitter Universe.Cosmin Crucean: Perturbative QED on de Sitter Universe.
Cosmin Crucean: Perturbative QED on de Sitter Universe.SEENET-MTP
 
Regression Theory
Regression TheoryRegression Theory
Regression TheorySSA KPI
 
Admissions in india 2015
Admissions in india 2015Admissions in india 2015
Admissions in india 2015Edhole.com
 
Scientific Computing with Python Webinar 9/18/2009:Curve Fitting
Scientific Computing with Python Webinar 9/18/2009:Curve FittingScientific Computing with Python Webinar 9/18/2009:Curve Fitting
Scientific Computing with Python Webinar 9/18/2009:Curve FittingEnthought, Inc.
 
Signal Processing Course : Sparse Regularization of Inverse Problems
Signal Processing Course : Sparse Regularization of Inverse ProblemsSignal Processing Course : Sparse Regularization of Inverse Problems
Signal Processing Course : Sparse Regularization of Inverse ProblemsGabriel Peyré
 
Andreas Eberle
Andreas EberleAndreas Eberle
Andreas EberleBigMC
 
Journey to structure from motion
Journey to structure from motionJourney to structure from motion
Journey to structure from motionJa-Keoung Koo
 
Gaussseidelsor
GaussseidelsorGaussseidelsor
Gaussseidelsoruis
 
4. standard granger causality
4. standard granger causality4. standard granger causality
4. standard granger causalityQuang Hoang
 
7. toda yamamoto-granger causality
7. toda yamamoto-granger causality7. toda yamamoto-granger causality
7. toda yamamoto-granger causalityQuang Hoang
 
Markov Tutorial CDC Shanghai 2009
Markov Tutorial CDC Shanghai 2009Markov Tutorial CDC Shanghai 2009
Markov Tutorial CDC Shanghai 2009Sean Meyn
 
กลศาสตร์เพิ่มเติม Ppt
กลศาสตร์เพิ่มเติม Pptกลศาสตร์เพิ่มเติม Ppt
กลศาสตร์เพิ่มเติม Ppttuiye
 
11.[104 111]analytical solution for telegraph equation by modified of sumudu ...
11.[104 111]analytical solution for telegraph equation by modified of sumudu ...11.[104 111]analytical solution for telegraph equation by modified of sumudu ...
11.[104 111]analytical solution for telegraph equation by modified of sumudu ...Alexander Decker
 
Computation of the marginal likelihood
Computation of the marginal likelihoodComputation of the marginal likelihood
Computation of the marginal likelihoodBigMC
 
Montpellier Math Colloquium
Montpellier Math ColloquiumMontpellier Math Colloquium
Montpellier Math ColloquiumChristian Robert
 
Introduction to inverse problems
Introduction to inverse problemsIntroduction to inverse problems
Introduction to inverse problemsDelta Pi Systems
 
Query Suggestion @ tokyotextmining#2
Query Suggestion @ tokyotextmining#2Query Suggestion @ tokyotextmining#2
Query Suggestion @ tokyotextmining#2ybenjo
 

What's hot (20)

修士論文発表会
修士論文発表会修士論文発表会
修士論文発表会
 
Cosmin Crucean: Perturbative QED on de Sitter Universe.
Cosmin Crucean: Perturbative QED on de Sitter Universe.Cosmin Crucean: Perturbative QED on de Sitter Universe.
Cosmin Crucean: Perturbative QED on de Sitter Universe.
 
Regression Theory
Regression TheoryRegression Theory
Regression Theory
 
Admissions in india 2015
Admissions in india 2015Admissions in india 2015
Admissions in india 2015
 
Scientific Computing with Python Webinar 9/18/2009:Curve Fitting
Scientific Computing with Python Webinar 9/18/2009:Curve FittingScientific Computing with Python Webinar 9/18/2009:Curve Fitting
Scientific Computing with Python Webinar 9/18/2009:Curve Fitting
 
Signal Processing Course : Sparse Regularization of Inverse Problems
Signal Processing Course : Sparse Regularization of Inverse ProblemsSignal Processing Course : Sparse Regularization of Inverse Problems
Signal Processing Course : Sparse Regularization of Inverse Problems
 
Andreas Eberle
Andreas EberleAndreas Eberle
Andreas Eberle
 
Journey to structure from motion
Journey to structure from motionJourney to structure from motion
Journey to structure from motion
 
Gaussseidelsor
GaussseidelsorGaussseidelsor
Gaussseidelsor
 
4. standard granger causality
4. standard granger causality4. standard granger causality
4. standard granger causality
 
7. toda yamamoto-granger causality
7. toda yamamoto-granger causality7. toda yamamoto-granger causality
7. toda yamamoto-granger causality
 
Markov Tutorial CDC Shanghai 2009
Markov Tutorial CDC Shanghai 2009Markov Tutorial CDC Shanghai 2009
Markov Tutorial CDC Shanghai 2009
 
กลศาสตร์เพิ่มเติม Ppt
กลศาสตร์เพิ่มเติม Pptกลศาสตร์เพิ่มเติม Ppt
กลศาสตร์เพิ่มเติม Ppt
 
11.[104 111]analytical solution for telegraph equation by modified of sumudu ...
11.[104 111]analytical solution for telegraph equation by modified of sumudu ...11.[104 111]analytical solution for telegraph equation by modified of sumudu ...
11.[104 111]analytical solution for telegraph equation by modified of sumudu ...
 
Computation of the marginal likelihood
Computation of the marginal likelihoodComputation of the marginal likelihood
Computation of the marginal likelihood
 
Cheat Sheet
Cheat SheetCheat Sheet
Cheat Sheet
 
Montpellier Math Colloquium
Montpellier Math ColloquiumMontpellier Math Colloquium
Montpellier Math Colloquium
 
Introduction to inverse problems
Introduction to inverse problemsIntroduction to inverse problems
Introduction to inverse problems
 
Query Suggestion @ tokyotextmining#2
Query Suggestion @ tokyotextmining#2Query Suggestion @ tokyotextmining#2
Query Suggestion @ tokyotextmining#2
 
iTute Notes MM
iTute Notes MMiTute Notes MM
iTute Notes MM
 

Viewers also liked

ICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM Prediction
ICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM PredictionICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM Prediction
ICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM PredictionSeiya Tokui
 
Overview of Chainer and Its Features
Overview of Chainer and Its FeaturesOverview of Chainer and Its Features
Overview of Chainer and Its FeaturesSeiya Tokui
 
今日からできる構造学習(主に構造化パーセプトロンについて)
今日からできる構造学習(主に構造化パーセプトロンについて)今日からできる構造学習(主に構造化パーセプトロンについて)
今日からできる構造学習(主に構造化パーセプトロンについて)syou6162
 
論文紹介 Compressing Neural Networks with the Hashing Trick
論文紹介 Compressing Neural Networks with the Hashing Trick論文紹介 Compressing Neural Networks with the Hashing Trick
論文紹介 Compressing Neural Networks with the Hashing TrickSeiya Tokui
 
深層学習フレームワークChainerの紹介とFPGAへの期待
深層学習フレームワークChainerの紹介とFPGAへの期待深層学習フレームワークChainerの紹介とFPGAへの期待
深層学習フレームワークChainerの紹介とFPGAへの期待Seiya Tokui
 
Introduction to Chainer
Introduction to ChainerIntroduction to Chainer
Introduction to ChainerSeiya Tokui
 
Differences of Deep Learning Frameworks
Differences of Deep Learning FrameworksDifferences of Deep Learning Frameworks
Differences of Deep Learning FrameworksSeiya Tokui
 
NIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding Model
NIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding ModelNIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding Model
NIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding ModelSeiya Tokui
 
Introduction to Chainer: A Flexible Framework for Deep Learning
Introduction to Chainer: A Flexible Framework for Deep LearningIntroduction to Chainer: A Flexible Framework for Deep Learning
Introduction to Chainer: A Flexible Framework for Deep LearningSeiya Tokui
 
Возможности сервиса
Возможности сервисаВозможности сервиса
Возможности сервисаcbwTest
 
2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau
2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau
2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im AutomobilbauVirtual Dimension Center (VDC) Fellbach
 
Guia de estudio metodologia de la investigacion se puede bajar universidad s...
Guia de estudio metodologia de la  investigacion se puede bajar universidad s...Guia de estudio metodologia de la  investigacion se puede bajar universidad s...
Guia de estudio metodologia de la investigacion se puede bajar universidad s...carlospadhz
 
Intervenciones grupales positivas para los alumnos con tdah
Intervenciones grupales positivas para los alumnos con tdahIntervenciones grupales positivas para los alumnos con tdah
Intervenciones grupales positivas para los alumnos con tdahFundación CADAH TDAH
 
Google Analytics as Database of Record
Google Analytics as Database of RecordGoogle Analytics as Database of Record
Google Analytics as Database of RecordDavy Tollenaere
 
How to prepare for an oui trial
How to prepare for an oui trialHow to prepare for an oui trial
How to prepare for an oui trialDavid Matson
 
Getting Started With Virtualization
Getting Started With VirtualizationGetting Started With Virtualization
Getting Started With VirtualizationBill Kalarness
 
Sesión 1 introdución online
Sesión 1 introdución onlineSesión 1 introdución online
Sesión 1 introdución onlineIsmael Bermúdez
 

Viewers also liked (20)

Ml4nlp 4 2
Ml4nlp 4 2Ml4nlp 4 2
Ml4nlp 4 2
 
ICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM Prediction
ICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM PredictionICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM Prediction
ICML2013読み会 Local Deep Kernel Learning for Efficient Non-linear SVM Prediction
 
Overview of Chainer and Its Features
Overview of Chainer and Its FeaturesOverview of Chainer and Its Features
Overview of Chainer and Its Features
 
今日からできる構造学習(主に構造化パーセプトロンについて)
今日からできる構造学習(主に構造化パーセプトロンについて)今日からできる構造学習(主に構造化パーセプトロンについて)
今日からできる構造学習(主に構造化パーセプトロンについて)
 
論文紹介 Compressing Neural Networks with the Hashing Trick
論文紹介 Compressing Neural Networks with the Hashing Trick論文紹介 Compressing Neural Networks with the Hashing Trick
論文紹介 Compressing Neural Networks with the Hashing Trick
 
深層学習フレームワークChainerの紹介とFPGAへの期待
深層学習フレームワークChainerの紹介とFPGAへの期待深層学習フレームワークChainerの紹介とFPGAへの期待
深層学習フレームワークChainerの紹介とFPGAへの期待
 
Introduction to Chainer
Introduction to ChainerIntroduction to Chainer
Introduction to Chainer
 
Differences of Deep Learning Frameworks
Differences of Deep Learning FrameworksDifferences of Deep Learning Frameworks
Differences of Deep Learning Frameworks
 
NIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding Model
NIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding ModelNIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding Model
NIPS2013読み会 DeViSE: A Deep Visual-Semantic Embedding Model
 
Introduction to Chainer: A Flexible Framework for Deep Learning
Introduction to Chainer: A Flexible Framework for Deep LearningIntroduction to Chainer: A Flexible Framework for Deep Learning
Introduction to Chainer: A Flexible Framework for Deep Learning
 
Возможности сервиса
Возможности сервисаВозможности сервиса
Возможности сервиса
 
2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau
2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau
2016-08 PM VDC veröffentlicht Whitepaper Virtuelle Techniken im Automobilbau
 
Guia de estudio metodologia de la investigacion se puede bajar universidad s...
Guia de estudio metodologia de la  investigacion se puede bajar universidad s...Guia de estudio metodologia de la  investigacion se puede bajar universidad s...
Guia de estudio metodologia de la investigacion se puede bajar universidad s...
 
Intervenciones grupales positivas para los alumnos con tdah
Intervenciones grupales positivas para los alumnos con tdahIntervenciones grupales positivas para los alumnos con tdah
Intervenciones grupales positivas para los alumnos con tdah
 
Google Analytics as Database of Record
Google Analytics as Database of RecordGoogle Analytics as Database of Record
Google Analytics as Database of Record
 
How to prepare for an oui trial
How to prepare for an oui trialHow to prepare for an oui trial
How to prepare for an oui trial
 
พิชิตชัยชาญ วรรณบุตร
พิชิตชัยชาญ  วรรณบุตรพิชิตชัยชาญ  วรรณบุตร
พิชิตชัยชาญ วรรณบุตร
 
Getting Started With Virtualization
Getting Started With VirtualizationGetting Started With Virtualization
Getting Started With Virtualization
 
Sesión 1 introdución online
Sesión 1 introdución onlineSesión 1 introdución online
Sesión 1 introdución online
 
Pretotyping primer 4
Pretotyping primer   4Pretotyping primer   4
Pretotyping primer 4
 

Similar to rinko2010

CVPR2010: higher order models in computer vision: Part 1, 2
CVPR2010: higher order models in computer vision: Part 1, 2CVPR2010: higher order models in computer vision: Part 1, 2
CVPR2010: higher order models in computer vision: Part 1, 2zukun
 
The dual geometry of Shannon information
The dual geometry of Shannon informationThe dual geometry of Shannon information
The dual geometry of Shannon informationFrank Nielsen
 
Lattices and codes
Lattices and codesLattices and codes
Lattices and codesSpringer
 
Intro probability 2
Intro probability 2Intro probability 2
Intro probability 2Phong Vo
 
Ml mle_bayes
Ml  mle_bayesMl  mle_bayes
Ml mle_bayesPhong Vo
 
Hybrid Atlas Models of Financial Equity Market
Hybrid Atlas Models of Financial Equity MarketHybrid Atlas Models of Financial Equity Market
Hybrid Atlas Models of Financial Equity Markettomoyukiichiba
 
March12 natarajan
March12 natarajanMarch12 natarajan
March12 natarajanBBKuhn
 
On Twisted Paraproducts and some other Multilinear Singular Integrals
On Twisted Paraproducts and some other Multilinear Singular IntegralsOn Twisted Paraproducts and some other Multilinear Singular Integrals
On Twisted Paraproducts and some other Multilinear Singular IntegralsVjekoslavKovac1
 
Additional notes EC220
Additional notes EC220Additional notes EC220
Additional notes EC220Guo Xu
 
Chapter2: Likelihood-based approach
Chapter2: Likelihood-based approach Chapter2: Likelihood-based approach
Chapter2: Likelihood-based approach Jae-kwang Kim
 
Introduction to Stochastic calculus
Introduction to Stochastic calculusIntroduction to Stochastic calculus
Introduction to Stochastic calculusAshwin Rao
 
584 fundamental theorem of calculus
584 fundamental theorem of calculus584 fundamental theorem of calculus
584 fundamental theorem of calculusgoldenratio618
 
Meta-learning and the ELBO
Meta-learning and the ELBOMeta-learning and the ELBO
Meta-learning and the ELBOYoonho Lee
 
Introduction to modern Variational Inference.
Introduction to modern Variational Inference.Introduction to modern Variational Inference.
Introduction to modern Variational Inference.Tomasz Kusmierczyk
 
Jyokyo-kai-20120605
Jyokyo-kai-20120605Jyokyo-kai-20120605
Jyokyo-kai-20120605ketanaka
 
EM algorithm and its application in probabilistic latent semantic analysis
EM algorithm and its application in probabilistic latent semantic analysisEM algorithm and its application in probabilistic latent semantic analysis
EM algorithm and its application in probabilistic latent semantic analysiszukun
 
Intro probability 3
Intro probability 3Intro probability 3
Intro probability 3Phong Vo
 

Similar to rinko2010 (20)

CVPR2010: higher order models in computer vision: Part 1, 2
CVPR2010: higher order models in computer vision: Part 1, 2CVPR2010: higher order models in computer vision: Part 1, 2
CVPR2010: higher order models in computer vision: Part 1, 2
 
The dual geometry of Shannon information
The dual geometry of Shannon informationThe dual geometry of Shannon information
The dual geometry of Shannon information
 
Lattices and codes
Lattices and codesLattices and codes
Lattices and codes
 
Intro probability 2
Intro probability 2Intro probability 2
Intro probability 2
 
Ml mle_bayes
Ml  mle_bayesMl  mle_bayes
Ml mle_bayes
 
ma112011id535
ma112011id535ma112011id535
ma112011id535
 
Hybrid Atlas Models of Financial Equity Market
Hybrid Atlas Models of Financial Equity MarketHybrid Atlas Models of Financial Equity Market
Hybrid Atlas Models of Financial Equity Market
 
March12 natarajan
March12 natarajanMarch12 natarajan
March12 natarajan
 
On Twisted Paraproducts and some other Multilinear Singular Integrals
On Twisted Paraproducts and some other Multilinear Singular IntegralsOn Twisted Paraproducts and some other Multilinear Singular Integrals
On Twisted Paraproducts and some other Multilinear Singular Integrals
 
Additional notes EC220
Additional notes EC220Additional notes EC220
Additional notes EC220
 
Chapter2: Likelihood-based approach
Chapter2: Likelihood-based approach Chapter2: Likelihood-based approach
Chapter2: Likelihood-based approach
 
Introduction to Stochastic calculus
Introduction to Stochastic calculusIntroduction to Stochastic calculus
Introduction to Stochastic calculus
 
584 fundamental theorem of calculus
584 fundamental theorem of calculus584 fundamental theorem of calculus
584 fundamental theorem of calculus
 
Random Forest
Random ForestRandom Forest
Random Forest
 
Meta-learning and the ELBO
Meta-learning and the ELBOMeta-learning and the ELBO
Meta-learning and the ELBO
 
Introduction to modern Variational Inference.
Introduction to modern Variational Inference.Introduction to modern Variational Inference.
Introduction to modern Variational Inference.
 
Jyokyo-kai-20120605
Jyokyo-kai-20120605Jyokyo-kai-20120605
Jyokyo-kai-20120605
 
EM algorithm and its application in probabilistic latent semantic analysis
EM algorithm and its application in probabilistic latent semantic analysisEM algorithm and its application in probabilistic latent semantic analysis
EM algorithm and its application in probabilistic latent semantic analysis
 
Slides mc gill-v3
Slides mc gill-v3Slides mc gill-v3
Slides mc gill-v3
 
Intro probability 3
Intro probability 3Intro probability 3
Intro probability 3
 

More from Seiya Tokui

Chainer/CuPy v5 and Future (Japanese)
Chainer/CuPy v5 and Future (Japanese)Chainer/CuPy v5 and Future (Japanese)
Chainer/CuPy v5 and Future (Japanese)Seiya Tokui
 
Chainer v2 and future dev plan
Chainer v2 and future dev planChainer v2 and future dev plan
Chainer v2 and future dev planSeiya Tokui
 
Chainer v2 alpha
Chainer v2 alphaChainer v2 alpha
Chainer v2 alphaSeiya Tokui
 
Learning stochastic neural networks with Chainer
Learning stochastic neural networks with ChainerLearning stochastic neural networks with Chainer
Learning stochastic neural networks with ChainerSeiya Tokui
 
深層学習フレームワーク Chainer の開発と今後の展開
深層学習フレームワーク Chainer の開発と今後の展開深層学習フレームワーク Chainer の開発と今後の展開
深層学習フレームワーク Chainer の開発と今後の展開Seiya Tokui
 
論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural Networks論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural NetworksSeiya Tokui
 
Chainer Update v1.8.0 -> v1.10.0+
Chainer Update v1.8.0 -> v1.10.0+Chainer Update v1.8.0 -> v1.10.0+
Chainer Update v1.8.0 -> v1.10.0+Seiya Tokui
 
生成モデルの Deep Learning
生成モデルの Deep Learning生成モデルの Deep Learning
生成モデルの Deep LearningSeiya Tokui
 
Chainer Development Plan 2015/12
Chainer Development Plan 2015/12Chainer Development Plan 2015/12
Chainer Development Plan 2015/12Seiya Tokui
 
Towards Chainer v1.5
Towards Chainer v1.5Towards Chainer v1.5
Towards Chainer v1.5Seiya Tokui
 
Deep Learningの基礎と応用
Deep Learningの基礎と応用Deep Learningの基礎と応用
Deep Learningの基礎と応用Seiya Tokui
 
Chainerの使い方と自然言語処理への応用
Chainerの使い方と自然言語処理への応用Chainerの使い方と自然言語処理への応用
Chainerの使い方と自然言語処理への応用Seiya Tokui
 
論文紹介 Semi-supervised Learning with Deep Generative Models
論文紹介 Semi-supervised Learning with Deep Generative Models論文紹介 Semi-supervised Learning with Deep Generative Models
論文紹介 Semi-supervised Learning with Deep Generative ModelsSeiya Tokui
 
Recurrent Neural Networks
Recurrent Neural NetworksRecurrent Neural Networks
Recurrent Neural NetworksSeiya Tokui
 
Deep learning実装の基礎と実践
Deep learning実装の基礎と実践Deep learning実装の基礎と実践
Deep learning実装の基礎と実践Seiya Tokui
 
Deep Learning技術の今
Deep Learning技術の今Deep Learning技術の今
Deep Learning技術の今Seiya Tokui
 
Deep Learningの技術と未来
Deep Learningの技術と未来Deep Learningの技術と未来
Deep Learningの技術と未来Seiya Tokui
 

More from Seiya Tokui (18)

Chainer/CuPy v5 and Future (Japanese)
Chainer/CuPy v5 and Future (Japanese)Chainer/CuPy v5 and Future (Japanese)
Chainer/CuPy v5 and Future (Japanese)
 
Chainer v3
Chainer v3Chainer v3
Chainer v3
 
Chainer v2 and future dev plan
Chainer v2 and future dev planChainer v2 and future dev plan
Chainer v2 and future dev plan
 
Chainer v2 alpha
Chainer v2 alphaChainer v2 alpha
Chainer v2 alpha
 
Learning stochastic neural networks with Chainer
Learning stochastic neural networks with ChainerLearning stochastic neural networks with Chainer
Learning stochastic neural networks with Chainer
 
深層学習フレームワーク Chainer の開発と今後の展開
深層学習フレームワーク Chainer の開発と今後の展開深層学習フレームワーク Chainer の開発と今後の展開
深層学習フレームワーク Chainer の開発と今後の展開
 
論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural Networks論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural Networks
 
Chainer Update v1.8.0 -> v1.10.0+
Chainer Update v1.8.0 -> v1.10.0+Chainer Update v1.8.0 -> v1.10.0+
Chainer Update v1.8.0 -> v1.10.0+
 
生成モデルの Deep Learning
生成モデルの Deep Learning生成モデルの Deep Learning
生成モデルの Deep Learning
 
Chainer Development Plan 2015/12
Chainer Development Plan 2015/12Chainer Development Plan 2015/12
Chainer Development Plan 2015/12
 
Towards Chainer v1.5
Towards Chainer v1.5Towards Chainer v1.5
Towards Chainer v1.5
 
Deep Learningの基礎と応用
Deep Learningの基礎と応用Deep Learningの基礎と応用
Deep Learningの基礎と応用
 
Chainerの使い方と自然言語処理への応用
Chainerの使い方と自然言語処理への応用Chainerの使い方と自然言語処理への応用
Chainerの使い方と自然言語処理への応用
 
論文紹介 Semi-supervised Learning with Deep Generative Models
論文紹介 Semi-supervised Learning with Deep Generative Models論文紹介 Semi-supervised Learning with Deep Generative Models
論文紹介 Semi-supervised Learning with Deep Generative Models
 
Recurrent Neural Networks
Recurrent Neural NetworksRecurrent Neural Networks
Recurrent Neural Networks
 
Deep learning実装の基礎と実践
Deep learning実装の基礎と実践Deep learning実装の基礎と実践
Deep learning実装の基礎と実践
 
Deep Learning技術の今
Deep Learning技術の今Deep Learning技術の今
Deep Learning技術の今
 
Deep Learningの技術と未来
Deep Learningの技術と未来Deep Learningの技術と未来
Deep Learningの技術と未来
 

Recently uploaded

How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhisoniya singh
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGSujit Pal
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...HostedbyConfluent
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...gurkirankumar98700
 

Recently uploaded (20)

How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAG
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
 

rinko2010

  • 1. CRF 2010 12 10 10 1 ( )
  • 3. 4 [Lafferty+, 01] Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data. John Lafferty, Andrew McCallum, Fernando Pereira. Proceedings of ICML’01, 2001. [Collins, 02] Discriminative training methods for hidden markov models: Theory and experiments with perceptron algorithms. Michael Collins. Proceedings of EMNLP’02, 2002. [Morency+, 07] Latent-dynamic discriminative models for continuous gesture recognition. Louis-Philippe Morency, Ariadna Quattoni, and Trevor Darrell. Proceedings of CVPR’07, 2007. [Sun+, 09] Latent Variable Perceptron Algorithm for Structured Classification. Xu Sun, Takuya Matsuzaki, Daisuke Okanohara and Jun’ichi Tsujii. Proceedings of IJCAI’09, 2009 3
  • 4. ‣ ‣ CRF Structured (Conditional Random Field) Perceptron 1 2 3 4 DPLVM Latent Variable (Discriminative Probabilistic Latent Variable Model) Perceptron ‣ 4
  • 5. x= x1 x2 xm y= y1 y2 ym y1 , . . . , ym ∈ Y 5
  • 6. ( : NP-chunking) x1 x2 x3 x4 x5 He is her brother . B O B I O y1 y2 y3 y4 y5 Y = {B, I, O} 6
  • 7. ‣ ‣ CRF Structured (Conditional Random Field) Perceptron 1 2 3 4 DPLVM Latent Variable (Discriminative Probabilistic Latent Variable Model) Perceptron ‣ 7
  • 8. Θ P (y|x, Θ) P (yi |xi , Θ) ∗ Θ {(xi , yi )}i=1 ∗ d .. . .. . d 8
  • 9. Θ P (y|x, Θ) x ˆ y = argmax P (y|x, Θ) ˆ y 9
  • 10. (x, y) →     f1 (y, x) Θ1   f2 (y, x)     Θ2    . .   . .  n  . · .  = F (y|x, Θ)      . .   . .   .   .  fn (y, x) Θn = = f (y, x) Θ 10
  • 11. 1 P (y|x, Θ) = exp F (y|x, Θ) Z Z= exp F (y |x, Θ) y F (y|x, Θ) = f (y, x) · Θ 1/Z argmax P (y|x, Θ) = argmax F (y|x, Θ) y y 11
  • 12. 1 P (y|x, Θ) = exp F (y|x, Θ) Z Z= exp F (y |x, Θ) y F (y|x, Θ) = fO(|Yx) ) Θ (y, |m · 1/Z argmax P (y|x, Θ) = argmax F (y|x, Θ) y y 12
  • 13. CRF: Conditional Random Field (sequential) yj−1 yj s(j, x, yj ) t(j, x, yj−1 , yj ) ⇒ 13
  • 14. CRF: Conditional Random Field (sequential) yj−1 yj s(j, x, yj ) t(j, x, yj−1 , yj ) ⇒ 14
  • 15. CRF d maximize log P (yi |xi , Θ) ∗ − R(Θ) i=1 R(Θ) Θ 15
  • 16. ‣ ‣ CRF Structured (Conditional Random Field) Perceptron 1 2 3 4 DPLVM Latent Variable (Discriminative Probabilistic Latent Variable Model) Perceptron ‣ 16
  • 17. Structured Perceptron ‣ ‣ (xi , yi ) ∗ F (yi |xi , Θ) ∗ =Θ· f (yi , xi ) ∗ (xi , yi ) ∗ yi = argmax F (y|xi , Θ ) i y yi = ∗ yi yi = ∗ yi Θ i+1 =Θ + i f (yi , xi ) ∗ − f (yi , xi ) Θ i+1 =Θ i 17
  • 18. Structured Perceptron Θ i+1 =Θ + i f (yi , xi ) ∗ − f (yi , xi ) Θ i+1 · (f (yi , xi ) ∗ − f (yi , xi )) 2 =Θ · i (f (yi , xi ) ∗ − f (yi , xi )) + f (yi , xi ) ∗ − f (yi , xi )2 ⇔ F (yi |xi , Θ ) ∗ i+1 − F (yi |xi , Θ i+1 ) 2 = F (yi |xi , Θi ) ∗ − F (yi |xi , Θ ) + i f (yi , xi ) ∗ − f (yi , xi )2 ≥0 18
  • 19. Structured Perceptron Θ i+1 =Θ + i f (yi , xi ) ∗ − f (yi , xi ) ∗ yi yi F (yi |xi , Θ ) ∗ i+1 − F (yi |xi , Θ i+1 ) 2 = F (yi |xi , Θi ) ∗ − F (yi |xi , Θ ) + i f (yi , xi ) ∗ − f (yi , xi )2 ≥0 19
  • 21. separability G(xi ) = {all possible label sequences for an example xi }, G(xi ) = G(xi ) − ∗ {yi } {(xi , yi )}d ∗ i=1 δ0 U2 = 1 U ∀i, ∀z ∈ G(xi ), F (yi |xi , U) − F (z|xi , U) ≥ δ. ∗ 21
  • 22. mistake bound δ0 {(xi , yi )}d ∗ i=1 M 2 R M≤ 2 δ R ∀i, ∀z ∈ G(xi ), f (yi , xi ) − f (z, xi )2 ≤ R ∗ d 22
  • 23. ‣ ‣ CRF Structured (Conditional Random Field) Perceptron 1 2 3 4 DPLVM Latent Variable (Discriminative Probabilistic Latent Variable Model) Perceptron ‣ 23
  • 24. They are her flowers . B O B I O They gave her flowers . B O B B O 24
  • 25. They are her flowers . B O B I O B1 They gave her flowers . B O B B O B2 25
  • 26. DPLVM - Discriminative Probabilistic Latent Variable Model Y ={ B , I , O }                    HB = { B1 , . . . , B|HB | } |HB | 26
  • 27. DPLVM - Discriminative Probabilistic Latent Variable Model y= y1 y2 ym h= h1 h2 hm ∀j, hj ∈ Hyj def. ⇐⇒ Proj(h) = y 27
  • 28. DPLVM (x, h) →     f1 (h, x) Θ1  f2 (h, x)   Θ2       . .   . .   . · .  = F (h|x, Θ)      .   .   . .   . .  fn (h, x) Θn = = f (h, x) Θ 28
  • 29. DPLVM h 1 P (h|x, Θ) = exp F (h|x, Θ) Z Z= exp F (h |x, Θ) h F (h|x, Θ) = f (h, x) · Θ f (h, x) argmax P (h|x, Θ) = argmax F (h|x, Θ) h h 29
  • 30. DPLVM ∗ (xi , yi ) h P (h|x, Θ) ∗ yi h P (y|x, Θ) = P (y|h, x, Θ)P (h|x, Θ) h = P (h|x, Θ) h:Proj(h)=y 30
  • 31. DPLVM d maximize log P (yi |xi , Θ) ∗ − R(Θ) i=1 R(Θ) Θ 31
  • 32. ‣ ‣ CRF Structured (Conditional Random Field) Perceptron 1 2 3 4 DPLVM Latent Variable (Discriminative Probabilistic Latent Variable Model) Perceptron ‣ 32
  • 33. Latent Variable Perceptron (xi , yi ) ∗ hi = argmax F (hi |xi , Θ), h yi = Proj(hi ) yi = ∗ yi yi = ∗ yi Θ i+1 =Θ +i f (hi , xi ) ∗ − f (h, xi ) Θ i+1 =Θ i ∗ hi ∗ hi = argmax F (h|xi , Θ ) i ∗ h:Proj(h)=yi 33
  • 34. mistake bound δ0 {(xi , yi )}i=1 ∗ d M 2T M 2 2 M≤ δ2 T d M = max f (y, xi )2 . i,y 34
  • 35. ‣ ‣ CRF Structured (Conditional Random Field) Perceptron 1 2 3 4 DPLVM Latent Variable (Discriminative Probabilistic Latent Variable Model) Perceptron ‣ 35
  • 36. ( ) ‣ X = {a, b} ‣ Y = {A, B} ‣ HA = {A1 , A2 }, HB = {B1 , B2 } ‣ P (hj |hj−1 ) P (xj |hj ) h x ‣ y = Proj(h) ‣ {(xi , yi )}i=1 ∗ d ‣ 36
  • 37. ( ) ‣ p from to A1 A2 B1 B2 A1 (1 − p)/3 (1 − p)/3 (1 − p)/3 p A2 p (1 − p)/3 (1 − p)/3 (1 − p)/3 B1 (1 − p)/3 p (1 − p)/3 (1 − p)/3 B2 (1 − p)/3 (1 − p)/3 p (1 − p)/3 ‣ P (xi = a|hi ) hi = A1 hi = A2 hi = B1 hi = B2 0.1 0.7 0.7 0.6 37
  • 38. ( ) Latent Variable Perceptron Structured Perceptron 100 90 accuracy [%] 80 70 60 50 40 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 p 38
  • 40. ‣ ‣ 40