SlideShare ist ein Scribd-Unternehmen logo
1 von 52
@koh_t
today’s menu
Machine Learning ?
Machine Learning ?
Machine Learning ?




             ,       ,
   ???
Machine Learning ?




   ...
Machine Learning ?




   ... !!
Machine Learning ?
Machine Learning ?




        = {       ,       ,   }


        ={    ,       ,           }
for an example...
Q.             ? (       )
for an example...
    Q.                        ? (   )

                                        …?



PC: he is ... ... ...

                   …

     id: Kan            id: poppo
 PC: he is Naoto Kan!
for another example...
Q.                        ?
for another example...
    Q.                         ?

                                   …?




PC: ... ...
PC: 1$ =      85 ± 0.25
many examples...
many examples...


…
many examples...


…
Quotes                                                    Daily Data




                                                                                   240
            130
            110
IBM

            90




                                                                                   230
            70
            200




                                                                                   220
Apple

            100




                                                                               E
            50




                                                                                   210
            35
            30
Microsoft

            25




                                                                                   200
            20




                                                                                                           13:00
            40 15




                                                                                   190
            30
Dell

            20
            10




                                                                                                           Hour
                    2005   2006       2007            2008   2009   2010

                                               Index



                                  (     )



                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000




                                                                           0             200   400   600          800   1000
1   m
1
                     n×m
            X∈X ⊂R

n
1   m
1
                     n×m
            X∈X ⊂R
            n   m     ?
n
many examples...
many examples...



 ”             ”
many examples...



 ”             ”

       OK
X   A,B




X       f(X)
X   A,B
     CX = max p(Ck |X)
               k




                                X       f(X)


f (X) = argmin ||Y − f (X)||F
           f
X   A,B
     CX = max p(Ck |X)
               k




                                X       f(X)


f (X) = argmin ||Y − f (X)||F
           f
my examples...
my examples...


i) 1
ii) 1
iii)    -5[V], -5 5[V]   , +5[V]
my examples...


i) 1
ii) 1
iii)    -5[V], -5 5[V]         , +5[V]


ii)                  iii)
                         OK?
my bachelor examples...

          Monthly Data             Daily Data




                             240
    250
    240




                             230
    230




                             220
    220
E




                         E
                             210
    210




                             200
    200




          June 15th                  13:00
    190




                             190
              Day                     Hour




                         …
my bachelor examples...

          Daily Data
                                         hourly Data
    240




                           225
    230
    220




                           220
E
    210




                       E
                           215
    200




            13:00




                           210
    190




                                 13:00       13:30     14:00



                           205
             Hour


                                             10 Sec




                                         …
my bachelor examples...
                                  …


AR:   x = (x(1), . . . , x(t))T ∈ X ⊂ Rt
      a = (a(1), . . . , a(t))T ∈ A ⊂ Rt
      x(t + 1) = a x + N (0, σ )
                    T              2




 0                    t t+1
my bachelor examples...
                                                           …


AR:   x = (x(1), . . . , x(t))T ∈ X ⊂ Rt
                                ARIMA Example


      a = (a(1), . . . , a(t))T ∈ A ⊂ Rt
      230



                    observed value


      x(t + 1) = a x + N (0, σ )
                    predicted value
                    T
                    2*SE           2
      225
      220
      215
      210
      205




 0          0   5      10        15      t t+1
                                        20      25   30   35

                                      Time
my bachelor examples...
i)               ˜
                 X                    Xr


ii) AR         &

X             ar = g(Xr )
              x(t + 1) = ar x + N (0, σ )
              ˜            T           2

              |˜(t + 1) − x(t)| → Ck
               x          ˜
         Xr
     ˜
     X
my bachelor examples...
i)               ˜
                 X                     Xr


ii) AR         &

X             ar = g(Xr )
              x(t + 1) = ar x + N (0, σ )
              ˜            T           2

              |˜(t + 1) − x(t)| → Ck
               x          ˜
         Xr
     ˜
     X




                                   …
my bachelor examples...

          Monthly Data             Daily Data




                             240
    250
    240




                             230
    230




                             220
    220




                         E
E




                             210
    210




                             200
    200




          June 15th                  13:00
    190




                             190
              Day                     Hour
my bachelor examples...
                   …


FFT:
K-SVD:
my bachelor examples...
                                                                                                                                                                                   …

 !"#$%
 &'()*                           FFT:                                             !"#$%
                                                                                  &'()*
K.TAKEUCHI                       k-SVD(GH<GI<'JKL) k-SVD(GH<GI<'JKL)
+,-(
                                 K-SVD:    K.TAKEUCHI

                                       1
                                                                                 +,-(
                                                                                   2
                                                                                                                                    1                                     2
                 1.0




                                                                 1.0




                                                                                                                 !0.214
./                                                                               ./                                           MN6OPQRST k U                                             MNOPQRS


                                                                                                                 !0.216




                                                                                                                                                        !0.215
                 0.5




                                                                 0.5




01                                                                               01                                           DVWQXY.                                                   UV6'<+D

                                                                                                                 !0.218
                 0.0




                                                                 0.0




                                                                                                                                                        !0.220
             E




                                                             E




                                                                                                             E




                                                                                                                                                    E
2301                                                                             2301



                                                                                                                 !0.220
.4                                                                               .4
                                                                                                                              ex)FFT Z[]^,weblet                                       TMNRWX
                 !0.5




                                                                 !0.5




                                                                                                                 !0.222




                                                                                                                                                        !0.225
567                                                                              567                             !0.224

                                                                                                                              _`Z[abD weblet Z                                          YZ[]'<
                 !1.0




                                                                 !1.0




                        0   20   40          60   80   100              0   20   40          60   80   100                5    10       15     20                5   10       15   20

89                                                                               89
                                                                                                                              RS
                                      Time                                            Time                                      Time                                  Time


k-NN                                   3                                         k-NN
                                                                                    4                                               3                                     4             Q_`XDaO
Local AR
                                                                                                                              k-SVD [cdDVWQe
                                                                                                                 !0.210




                                                                                 Local AR




                                                                                                                                                        !0.205
                                                                                                                                                                                        ex) fgh, ij
                 1.0




                                                                 1.0




k-SVD                                                                            k-SVD




                                                                                                                                                        !0.210
                                                                                                                              fghi
                 0.5




                                                                 0.5




                                                                                                                 !0.215




:;'<+                                                                            :;'<+



                                                                                                                                                        !0.215
                                                                                                             E




                                                                                                                                                    E
                 0.0




                                                                 0.0
             E




                                                             E




                                                                                                                 !0.220




5=>?                                                                             5=>?

                                                                                                                                                        !0.220
                 !0.5




                                                                 !0.5




                                                                                                                                                        !0.225
@A                                                                               @A
                                                                                                                 !0.225
                 !1.0




                                                                 !1.0




                                                                                 BCDEF
                                                                                                                          5    10       15     20                5   10       15   20

BCDEF                   0   20   40

                                      Time
                                             60   80   100              0   20   40

                                                                                      Time
                                                                                             60   80   100
                                                                                                                                Time                                  Time




                                                         FFT
                                                       FFT                                                                                   K-SVD
                                                                                                                                             Dictionary
my bachelor examples...
                                       K-SVD:
                                       i)                                                       D

                                            argmin ||X −                            2
                                                                                DZ||F     s.t. ∀i ||zi || ≤ C0
                                                         D,Z

                                       ii)                                            ˜
                                                                                      X         D
$%
)*

UCHI
                     k-SVD(GH<GI<'JKL)
                              1                                   2


                                                                                |˜(t − 1) − x(t)| k-SVD(GH<GI<'JKL)
                                                                                 x          ˜
                                                                                            !"#$%
                                                                                            &'()*
                                                                                                   → Ck
            !0.214




                                                                                MNOPQRST   K.TAKEUCHI
            !0.216




                                                             ×
                                                !0.215




                                                                                UV6'<+DPQRS
            !0.218




                                                                                       +,-(                                     k!SVD regression
                                                !0.220
        E




                                            E
            !0.220




                                                                                TMNRWX ./                                                                         MNOPQRSTUV
            !0.222




                                                !0.225




                                                                                       01                                                                         WX
            !0.224




                                                                                YZ[]'<+^SP

                                                                                                            200
                                                                                           2301
                                                                                           .4                                                                     k YDPQDZ[+
    D
                     5   10       15   20                5   10       15   20

                          Time                                Time



                              3                                   4             Q_`XDaObcde
                                                                                           567
                                                                                                                                                                  ]^-_`aObPQ
                                                                                                            150
                                                                                           89
                                                                                                                                                                  YcdVef
            !0.210




                                                                                                        E
                                                !0.205




                                                                                ex) fgh, ij, kl
                                                                                           k-NN
                                                                                           Local AR
                                                                                                                                                                  PQU l+1 RgXV
                                                !0.210
            !0.215




                                                                                           k-SVD
+
                                                                                                                                                                  X!_h 1-l Dijk
                                                                                                            100
                                                !0.215




                                                                                           :;'<+
        E




                                            E




                                                                                                                                                                  ma+<]^-n`
            !0.220




                                                !0.220




                                                                                           5=>?

                                                                                                                                                                  ObPQUopqrs
                                                !0.225




                                                                                           @A
                                                                                                            50
            !0.225




                                                                                                                                                                  VWXSTD l+1 Di
                                                                                                                  0   5   10         15            20   25   30


                                                                                           BCDEF                                      Time


F
                     5   10       15   20                5   10       15   20                                                       k=60,T=18

                          Time                                Time


                                                                                                                               k-SVD                              Utuvnwaxyz
my bachelor examples...
                                       K-SVD:
                                       i)                                                       D

                                            argmin ||X −                            2
                                                                                DZ||F     s.t. ∀i ||zi || ≤ C0
                                                         D,Z

                                       ii)                                            ˜
                                                                                      X         D
$%
)*

UCHI
                     k-SVD(GH<GI<'JKL)
                              1                                   2


                                                                                |˜(t − 1) − x(t)| k-SVD(GH<GI<'JKL)
                                                                                 x          ˜
                                                                                            !"#$%
                                                                                            &'()*
                                                                                                   → Ck
            !0.214




                                                                                MNOPQRST   K.TAKEUCHI
            !0.216




                                                             ×
                                                !0.215




                                                                                UV6'<+DPQRS
            !0.218




                                                                                       +,-(                                     k!SVD regression
                                                !0.220
        E




                                            E
            !0.220




                                                                                TMNRWX ./                                                                         MNOPQRSTUV
            !0.222




                                                !0.225




                                                                                       01                                                                         WX
            !0.224




                                                                                YZ[]'<+^SP

                                                                                                            200
                                                                                           2301
                                                                                           .4                                                                     k YDPQDZ[+
    D
                     5   10       15   20                5   10       15   20

                          Time                                Time



                              3                                   4             Q_`XDaObcde
                                                                                           567
                                                                                                                                                                  ]^-_`aObPQ
                                                                                                            150
                                                                                           89
                                                                                                                                                                  YcdVef
            !0.210




                                                                                                        E
                                                !0.205




                                                                                ex) fgh, ij, kl
                                                                                           k-NN
                                                                                           Local AR
                                                                                                                                                                  PQU l+1 RgXV
                                                !0.210
            !0.215




                                                                                           k-SVD
+
                                                                                                                                                                  X!_h 1-l Dijk
                                                                                                            100
                                                !0.215




                                                                                           :;'<+
        E




                                            E




                                                                                                                                                                  ma+<]^-n`
            !0.220




                                                !0.220




                                                                                           5=>?

                                                                                                                                                                  ObPQUopqrs
                                                !0.225




                                                                                           @A
                                                                                                            50
            !0.225




                                                                                                                                                                  VWXSTD l+1 Di
                                                                                                                  0   5   10         15            20   25   30


                                                                                           BCDEF                                      Time


F
                     5   10       15   20                5   10       15   20                                                       k=60,T=18

                          Time                                Time


                                                                                                                               k-SVD                              Utuvnwaxyz
my examples...
my examples...


K-SVD
my examples...


K-SVD
my examples...


K-SVD




                  cf: no free lunch theorem
@koh_t

Weitere ähnliche Inhalte

Was ist angesagt? (6)

annalist - a statistics aggregator written in erlang
annalist - a statistics aggregator written in erlangannalist - a statistics aggregator written in erlang
annalist - a statistics aggregator written in erlang
 
Line by environment interaction, yield stability and grouping of test locatio...
Line by environment interaction, yield stability and grouping of test locatio...Line by environment interaction, yield stability and grouping of test locatio...
Line by environment interaction, yield stability and grouping of test locatio...
 
AESC Perth 2008
AESC Perth 2008AESC Perth 2008
AESC Perth 2008
 
【東北大学】平成19年環境報告書
【東北大学】平成19年環境報告書【東北大学】平成19年環境報告書
【東北大学】平成19年環境報告書
 
Regulations As a "Panacea": Exploring the Consequences
Regulations As a "Panacea": Exploring the ConsequencesRegulations As a "Panacea": Exploring the Consequences
Regulations As a "Panacea": Exploring the Consequences
 
【森林総合研究所】平成18年環境報告
【森林総合研究所】平成18年環境報告【森林総合研究所】平成18年環境報告
【森林総合研究所】平成18年環境報告
 

Andere mochten auch

C:\fakepath\friends forever slideshow2
C:\fakepath\friends forever slideshow2C:\fakepath\friends forever slideshow2
C:\fakepath\friends forever slideshow2
xobrixo396
 
opening
openingopening
opening
koh-t
 
Socialism swissa
Socialism swissaSocialism swissa
Socialism swissa
swissa22
 
closing
closingclosing
closing
koh-t
 
Luciano pavarotti -_popular_italian_songs_(book)
Luciano pavarotti -_popular_italian_songs_(book)Luciano pavarotti -_popular_italian_songs_(book)
Luciano pavarotti -_popular_italian_songs_(book)
Daniel Clavero
 
Tomlin policies and procedures 2
Tomlin policies and procedures 2Tomlin policies and procedures 2
Tomlin policies and procedures 2
esimmons8
 
Médias sociaux : Votre marque ne vous appartient plus !
Médias sociaux : Votre marque ne vous appartient plus !Médias sociaux : Votre marque ne vous appartient plus !
Médias sociaux : Votre marque ne vous appartient plus !
Frederic CAVAZZA
 

Andere mochten auch (19)

Mmix m5-uf2-nf1
Mmix m5-uf2-nf1Mmix m5-uf2-nf1
Mmix m5-uf2-nf1
 
C:\fakepath\friends forever slideshow2
C:\fakepath\friends forever slideshow2C:\fakepath\friends forever slideshow2
C:\fakepath\friends forever slideshow2
 
Social Media For Business
Social Media For BusinessSocial Media For Business
Social Media For Business
 
Friends forever slideshow2
Friends forever slideshow2Friends forever slideshow2
Friends forever slideshow2
 
Mmix m5-uf2-nf2-nf3
Mmix m5-uf2-nf2-nf3Mmix m5-uf2-nf2-nf3
Mmix m5-uf2-nf2-nf3
 
Sustainability in the hotel industry.
Sustainability in the hotel industry.Sustainability in the hotel industry.
Sustainability in the hotel industry.
 
opening
openingopening
opening
 
Andf newsletter no.2
Andf newsletter no.2Andf newsletter no.2
Andf newsletter no.2
 
Socialism swissa
Socialism swissaSocialism swissa
Socialism swissa
 
closing
closingclosing
closing
 
Marine Fauna
Marine FaunaMarine Fauna
Marine Fauna
 
Luciano pavarotti -_popular_italian_songs_(book)
Luciano pavarotti -_popular_italian_songs_(book)Luciano pavarotti -_popular_italian_songs_(book)
Luciano pavarotti -_popular_italian_songs_(book)
 
Activities
ActivitiesActivities
Activities
 
The Eco Lodge
The Eco LodgeThe Eco Lodge
The Eco Lodge
 
ZipComputers
ZipComputersZipComputers
ZipComputers
 
In what way does adam smith & david
In what way does adam smith & davidIn what way does adam smith & david
In what way does adam smith & david
 
Tomlin policies and procedures 2
Tomlin policies and procedures 2Tomlin policies and procedures 2
Tomlin policies and procedures 2
 
Médias sociaux : Votre marque ne vous appartient plus !
Médias sociaux : Votre marque ne vous appartient plus !Médias sociaux : Votre marque ne vous appartient plus !
Médias sociaux : Votre marque ne vous appartient plus !
 
Initiation au couplage réalité augmentée (RA) - système d’information géograp...
Initiation au couplage réalité augmentée (RA) - système d’information géograp...Initiation au couplage réalité augmentée (RA) - système d’information géograp...
Initiation au couplage réalité augmentée (RA) - système d’information géograp...
 

Ähnlich wie slide

Scania HQ%20Bank%20Investor%20meeting_tcm10-219785
Scania HQ%20Bank%20Investor%20meeting_tcm10-219785Scania HQ%20Bank%20Investor%20meeting_tcm10-219785
Scania HQ%20Bank%20Investor%20meeting_tcm10-219785
finance50
 
Hr Achievement Gaphs Thru 2009 Rev
Hr Achievement Gaphs Thru 2009 RevHr Achievement Gaphs Thru 2009 Rev
Hr Achievement Gaphs Thru 2009 Rev
shg
 
2008 O Level Analysis
2008 O Level Analysis2008 O Level Analysis
2008 O Level Analysis
moe_maldives
 

Ähnlich wie slide (20)

Analyzing Probabilistic Models in Hierarchical BOA on Traps and Spin Glasses
Analyzing Probabilistic Models in Hierarchical BOA on Traps and Spin GlassesAnalyzing Probabilistic Models in Hierarchical BOA on Traps and Spin Glasses
Analyzing Probabilistic Models in Hierarchical BOA on Traps and Spin Glasses
 
Facebook: an investment for the future
Facebook: an investment for the futureFacebook: an investment for the future
Facebook: an investment for the future
 
From Technology to Product
From Technology to ProductFrom Technology to Product
From Technology to Product
 
Bast en 2011
Bast en 2011Bast en 2011
Bast en 2011
 
Scania HQ%20Bank%20Investor%20meeting_tcm10-219785
Scania HQ%20Bank%20Investor%20meeting_tcm10-219785Scania HQ%20Bank%20Investor%20meeting_tcm10-219785
Scania HQ%20Bank%20Investor%20meeting_tcm10-219785
 
Extending Io Scalability
Extending Io ScalabilityExtending Io Scalability
Extending Io Scalability
 
Tinker global energy transition feb 2012
Tinker global energy transition feb 2012Tinker global energy transition feb 2012
Tinker global energy transition feb 2012
 
SPICE MODEL of ERZV09D390 in SPICE PARK
SPICE MODEL of ERZV09D390 in SPICE PARKSPICE MODEL of ERZV09D390 in SPICE PARK
SPICE MODEL of ERZV09D390 in SPICE PARK
 
Pawel karas
Pawel karasPawel karas
Pawel karas
 
SPICE MODEL of TPCA8004-H (Standard+BDS Model) in SPICE PARK
SPICE MODEL of TPCA8004-H (Standard+BDS Model) in SPICE PARKSPICE MODEL of TPCA8004-H (Standard+BDS Model) in SPICE PARK
SPICE MODEL of TPCA8004-H (Standard+BDS Model) in SPICE PARK
 
SPICE MODEL of TPCA8004-H (Professional+BDP Model) in SPICE PARK
SPICE MODEL of TPCA8004-H (Professional+BDP Model) in SPICE PARKSPICE MODEL of TPCA8004-H (Professional+BDP Model) in SPICE PARK
SPICE MODEL of TPCA8004-H (Professional+BDP Model) in SPICE PARK
 
SPICE MODEL of TPCM8002-H (Standard+BDS Model) in SPICE PARK
SPICE MODEL of TPCM8002-H (Standard+BDS Model) in SPICE PARKSPICE MODEL of TPCM8002-H (Standard+BDS Model) in SPICE PARK
SPICE MODEL of TPCM8002-H (Standard+BDS Model) in SPICE PARK
 
Hr Achievement Gaphs Thru 2009 Rev
Hr Achievement Gaphs Thru 2009 RevHr Achievement Gaphs Thru 2009 Rev
Hr Achievement Gaphs Thru 2009 Rev
 
Hr Achievement Gaphs Thru 2009 Rev
Hr Achievement Gaphs Thru 2009 RevHr Achievement Gaphs Thru 2009 Rev
Hr Achievement Gaphs Thru 2009 Rev
 
SPICE MODEL of TPCA8011-H (Professional+BDP Model) in SPICE PARK
SPICE MODEL of TPCA8011-H (Professional+BDP Model) in SPICE PARKSPICE MODEL of TPCA8011-H (Professional+BDP Model) in SPICE PARK
SPICE MODEL of TPCA8011-H (Professional+BDP Model) in SPICE PARK
 
Fusion improvement
Fusion improvementFusion improvement
Fusion improvement
 
2008 O Level Analysis
2008 O Level Analysis2008 O Level Analysis
2008 O Level Analysis
 
SPICE MODEL of TPCM8002-H (Professional+BDP Model) in SPICE PARK
SPICE MODEL of TPCM8002-H (Professional+BDP Model) in SPICE PARKSPICE MODEL of TPCM8002-H (Professional+BDP Model) in SPICE PARK
SPICE MODEL of TPCM8002-H (Professional+BDP Model) in SPICE PARK
 
SPICE MODEL of ERZV09D270 in SPICE PARK
SPICE MODEL of ERZV09D270 in SPICE PARKSPICE MODEL of ERZV09D270 in SPICE PARK
SPICE MODEL of ERZV09D270 in SPICE PARK
 
SPICE MODEL of ERZV09D220 in SPICE PARK
SPICE MODEL of ERZV09D220 in SPICE PARKSPICE MODEL of ERZV09D220 in SPICE PARK
SPICE MODEL of ERZV09D220 in SPICE PARK
 

Kürzlich hochgeladen

Spellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please PractiseSpellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please Practise
AnaAcapella
 
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
ZurliaSoop
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
ciinovamais
 
Salient Features of India constitution especially power and functions
Salient Features of India constitution especially power and functionsSalient Features of India constitution especially power and functions
Salient Features of India constitution especially power and functions
KarakKing
 

Kürzlich hochgeladen (20)

Spatium Project Simulation student brief
Spatium Project Simulation student briefSpatium Project Simulation student brief
Spatium Project Simulation student brief
 
Making communications land - Are they received and understood as intended? we...
Making communications land - Are they received and understood as intended? we...Making communications land - Are they received and understood as intended? we...
Making communications land - Are they received and understood as intended? we...
 
Dyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptxDyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptx
 
FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024
 
Spellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please PractiseSpellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please Practise
 
Introduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The BasicsIntroduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The Basics
 
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
ICT role in 21st century education and it's challenges.
ICT role in 21st century education and it's challenges.ICT role in 21st century education and it's challenges.
ICT role in 21st century education and it's challenges.
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
Key note speaker Neum_Admir Softic_ENG.pdf
Key note speaker Neum_Admir Softic_ENG.pdfKey note speaker Neum_Admir Softic_ENG.pdf
Key note speaker Neum_Admir Softic_ENG.pdf
 
How to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POSHow to Manage Global Discount in Odoo 17 POS
How to Manage Global Discount in Odoo 17 POS
 
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
 
Sociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning ExhibitSociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning Exhibit
 
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptxBasic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
 
Unit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptxUnit-IV- Pharma. Marketing Channels.pptx
Unit-IV- Pharma. Marketing Channels.pptx
 
Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.ppt
 
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
 
This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.
 
Salient Features of India constitution especially power and functions
Salient Features of India constitution especially power and functionsSalient Features of India constitution especially power and functions
Salient Features of India constitution especially power and functions
 

slide

  • 9. Machine Learning ? = { , , } ={ , , }
  • 11. for an example... Q. ? ( ) …? PC: he is ... ... ... … id: Kan id: poppo PC: he is Naoto Kan!
  • 13. for another example... Q. ? …? PC: ... ... PC: 1$ = 85 ± 0.25
  • 17.
  • 18.
  • 19.
  • 20.
  • 21. Quotes Daily Data 240 130 110 IBM 90 230 70 200 220 Apple 100 E 50 210 35 30 Microsoft 25 200 20 13:00 40 15 190 30 Dell 20 10 Hour 2005 2006 2007 2008 2009 2010 Index ( ) 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000 0 200 400 600 800 1000
  • 22.
  • 23. 1 m 1 n×m X∈X ⊂R n
  • 24. 1 m 1 n×m X∈X ⊂R n m ? n
  • 28.
  • 29.
  • 30. X A,B X f(X)
  • 31. X A,B CX = max p(Ck |X) k X f(X) f (X) = argmin ||Y − f (X)||F f
  • 32. X A,B CX = max p(Ck |X) k X f(X) f (X) = argmin ||Y − f (X)||F f
  • 33.
  • 35. my examples... i) 1 ii) 1 iii) -5[V], -5 5[V] , +5[V]
  • 36. my examples... i) 1 ii) 1 iii) -5[V], -5 5[V] , +5[V] ii) iii) OK?
  • 37. my bachelor examples... Monthly Data Daily Data 240 250 240 230 230 220 220 E E 210 210 200 200 June 15th 13:00 190 190 Day Hour …
  • 38. my bachelor examples... Daily Data hourly Data 240 225 230 220 220 E 210 E 215 200 13:00 210 190 13:00 13:30 14:00 205 Hour 10 Sec …
  • 39. my bachelor examples... … AR: x = (x(1), . . . , x(t))T ∈ X ⊂ Rt a = (a(1), . . . , a(t))T ∈ A ⊂ Rt x(t + 1) = a x + N (0, σ ) T 2 0 t t+1
  • 40. my bachelor examples... … AR: x = (x(1), . . . , x(t))T ∈ X ⊂ Rt ARIMA Example a = (a(1), . . . , a(t))T ∈ A ⊂ Rt 230 observed value x(t + 1) = a x + N (0, σ ) predicted value T 2*SE 2 225 220 215 210 205 0 0 5 10 15 t t+1 20 25 30 35 Time
  • 41. my bachelor examples... i) ˜ X Xr ii) AR & X ar = g(Xr ) x(t + 1) = ar x + N (0, σ ) ˜ T 2 |˜(t + 1) − x(t)| → Ck x ˜ Xr ˜ X
  • 42. my bachelor examples... i) ˜ X Xr ii) AR & X ar = g(Xr ) x(t + 1) = ar x + N (0, σ ) ˜ T 2 |˜(t + 1) − x(t)| → Ck x ˜ Xr ˜ X …
  • 43. my bachelor examples... Monthly Data Daily Data 240 250 240 230 230 220 220 E E 210 210 200 200 June 15th 13:00 190 190 Day Hour
  • 44. my bachelor examples... … FFT: K-SVD:
  • 45. my bachelor examples... … !"#$% &'()* FFT: !"#$% &'()* K.TAKEUCHI k-SVD(GH<GI<'JKL) k-SVD(GH<GI<'JKL) +,-( K-SVD: K.TAKEUCHI 1 +,-( 2 1 2 1.0 1.0 !0.214 ./ ./ MN6OPQRST k U MNOPQRS !0.216 !0.215 0.5 0.5 01 01 DVWQXY. UV6'<+D !0.218 0.0 0.0 !0.220 E E E E 2301 2301 !0.220 .4 .4 ex)FFT Z[]^,weblet TMNRWX !0.5 !0.5 !0.222 !0.225 567 567 !0.224 _`Z[abD weblet Z YZ[]'< !1.0 !1.0 0 20 40 60 80 100 0 20 40 60 80 100 5 10 15 20 5 10 15 20 89 89 RS Time Time Time Time k-NN 3 k-NN 4 3 4 Q_`XDaO Local AR k-SVD [cdDVWQe !0.210 Local AR !0.205 ex) fgh, ij 1.0 1.0 k-SVD k-SVD !0.210 fghi 0.5 0.5 !0.215 :;'<+ :;'<+ !0.215 E E 0.0 0.0 E E !0.220 5=>? 5=>? !0.220 !0.5 !0.5 !0.225 @A @A !0.225 !1.0 !1.0 BCDEF 5 10 15 20 5 10 15 20 BCDEF 0 20 40 Time 60 80 100 0 20 40 Time 60 80 100 Time Time FFT FFT K-SVD Dictionary
  • 46. my bachelor examples... K-SVD: i) D argmin ||X − 2 DZ||F s.t. ∀i ||zi || ≤ C0 D,Z ii) ˜ X D $% )* UCHI k-SVD(GH<GI<'JKL) 1 2 |˜(t − 1) − x(t)| k-SVD(GH<GI<'JKL) x ˜ !"#$% &'()* → Ck !0.214 MNOPQRST K.TAKEUCHI !0.216 × !0.215 UV6'<+DPQRS !0.218 +,-( k!SVD regression !0.220 E E !0.220 TMNRWX ./ MNOPQRSTUV !0.222 !0.225 01 WX !0.224 YZ[]'<+^SP 200 2301 .4 k YDPQDZ[+ D 5 10 15 20 5 10 15 20 Time Time 3 4 Q_`XDaObcde 567 ]^-_`aObPQ 150 89 YcdVef !0.210 E !0.205 ex) fgh, ij, kl k-NN Local AR PQU l+1 RgXV !0.210 !0.215 k-SVD + X!_h 1-l Dijk 100 !0.215 :;'<+ E E ma+<]^-n` !0.220 !0.220 5=>? ObPQUopqrs !0.225 @A 50 !0.225 VWXSTD l+1 Di 0 5 10 15 20 25 30 BCDEF Time F 5 10 15 20 5 10 15 20 k=60,T=18 Time Time k-SVD Utuvnwaxyz
  • 47. my bachelor examples... K-SVD: i) D argmin ||X − 2 DZ||F s.t. ∀i ||zi || ≤ C0 D,Z ii) ˜ X D $% )* UCHI k-SVD(GH<GI<'JKL) 1 2 |˜(t − 1) − x(t)| k-SVD(GH<GI<'JKL) x ˜ !"#$% &'()* → Ck !0.214 MNOPQRST K.TAKEUCHI !0.216 × !0.215 UV6'<+DPQRS !0.218 +,-( k!SVD regression !0.220 E E !0.220 TMNRWX ./ MNOPQRSTUV !0.222 !0.225 01 WX !0.224 YZ[]'<+^SP 200 2301 .4 k YDPQDZ[+ D 5 10 15 20 5 10 15 20 Time Time 3 4 Q_`XDaObcde 567 ]^-_`aObPQ 150 89 YcdVef !0.210 E !0.205 ex) fgh, ij, kl k-NN Local AR PQU l+1 RgXV !0.210 !0.215 k-SVD + X!_h 1-l Dijk 100 !0.215 :;'<+ E E ma+<]^-n` !0.220 !0.220 5=>? ObPQUopqrs !0.225 @A 50 !0.225 VWXSTD l+1 Di 0 5 10 15 20 25 30 BCDEF Time F 5 10 15 20 5 10 15 20 k=60,T=18 Time Time k-SVD Utuvnwaxyz
  • 51. my examples... K-SVD cf: no free lunch theorem

Hinweis der Redaktion