機械学習と解釈可能性

Transcript

1. 1.

   ػցֶशͱղऍՄೳੑ Masaaki Horikoshi @ ARISE analytics

2. 2.

   ࣗݾ঺հ • R • ύοέʔδ։ൃͳͲ • Git Awards ࠃ಺1Ґ •

   Python • http://git-awards.com/users/search?login=sinhrks
3. 3.

   Interpretability ղऍՄೳੑ

4. 4.

   ղऍՄೳੑͱ͸ • ໌֬ͳఆٛ͸ͳ͍͕ • Ϟσϧͷग़ྗ (what) ͚ͩͰͳ͘ɺͦͷཧ༝ (why) Λઆ໌͢Δ͜ͱ •

   ͦͷͨΊʹɺͳΜΒ͔ͷख๏/ج४Λ༻͍ͯϞσϧΛղऍ͢Δ͜ͱ The Mythos of Model Interpretability (Lipton, 2016)
5. 5.

   ͳʹ͕خ͍͔͠ • Trust • Causality • Transferability • Informativeness •

   Fair and Ethical Decision Making The Mythos of Model Interpretability (Lipton, 2016)
6. 6.

   ղऍͷͨΊͷΞϓϩʔν 1. આ໌͠΍͍͢ػցֶशख๏ΛબͿ • ਫ਼౓͕ෆे෼ͳ৔߹͕͋Δ 2. ͳΜΒ͔ͷղऍख๏Λ࢖͏

7. 7.

   ղऍՄೳੑ • Global Interpretability • Ϟσϧ΍σʔλશମͷ܏޲Λղऍ • ۙࣅ΍ཁ໿౷ܭྔΛར༻ => ہॴతʹ͸ෆਖ਼֬ͳ৔߹΋

   • Local Interpretability • Ϟσϧ΍σʔλͷݶΒΕͨྖҬΛղऍ • ΑΓਖ਼֬ͳઆ໌͕Մೳ
8. 8.

   ղऍՄೳੑ • ద੾ͳख๏͸ʮԿΛʯղऍ͍͔ͨ͠ʹґଘ .PEFM
   4QFDJpD .PEFM
   "HOPTUJD (MPCBM
   *OUFSQSFUBCJMJUZ w 3FHSFTTJPO
   $PF⒏DJFOUT
   w

   'FBUVSF
   *NQPSUBODF
   ʜ w 4VSSPHBUF
   .PEFMT
   w 4FOTJUJWJUZ
   "OBMZTJT
   ʜ -PDBM
   *OUFSQSFUBCJMJUZ w .BYJNVN
   "DUJWBUJPO
   "OBMZTJT
   ʜ w -*.&
   w -0$0
   w 4)"1
   ʜ
9. 9.

   Regression Coefficients • ඪ४Խภճؼ܎਺ library(dplyr) library(mlbench) data(BostonHousing) df <- BostonHousing

   %>% mutate_if(is.factor, as.numeric) %>% mutate_at(-14, scale) head(df, 3) crim zn indus chas nox rm age 1 -0.4193669 0.2845483 -1.2866362 -0.2723291 -0.1440749 0.4132629 -0.1198948 2 -0.4169267 -0.4872402 -0.5927944 -0.2723291 -0.7395304 0.1940824 0.3668034 3 -0.4169290 -0.4872402 -0.5927944 -0.2723291 -0.7395304 1.2814456 -0.2655490 dis rad tax ptratio b lstat medv 1 0.140075 -0.9818712 -0.6659492 -1.4575580 0.4406159 -1.0744990 24.0 2 0.556609 -0.8670245 -0.9863534 -0.3027945 0.4406159 -0.4919525 21.6 3 0.556609 -0.8670245 -0.9863534 -0.3027945 0.3960351 -1.2075324 34.7
10. 10.

    Regression Coefficients • ඪ४Խภճؼ܎਺ • coefplot ύοέʔδͰภճؼ܎਺ͷՄࢹԽ ͕Մೳ lm.fit <-

    lm(medv ~ ., data = df) coef(lm.fit) (Intercept) crim zn indus chas nox rm 22.53280632 -0.92906457 1.08263896 0.14103943 0.68241438 -2.05875361 2.67687661 age dis rad tax ptratio b lstat 0.01948534 -3.10711605 2.66485220 -2.07883689 -2.06264585 0.85010886 -3.74733185
11. 11.

    • coefplot Regression Coefficients library(coefplot) coefplot(lm.fit) σʔλͷෆඋΛ౷ܭతʹݟൈ͘ (Gelman's Secret Weapon)

    http://d.hatena.ne.jp/hoxo_m/20150617/p1
12. 12.

    • Feature Importance (Random Forest) Feature Importance library(dplyr) library(mlbench) data(Sonar)

    head(Sonar, 3) V1 V2 V3 V4 V5 V6 V7 V8 V9 1 0.0200 0.0371 0.0428 0.0207 0.0954 0.0986 0.1539 0.1601 0.3109 2 0.0453 0.0523 0.0843 0.0689 0.1183 0.2583 0.2156 0.3481 0.3337 3 0.0262 0.0582 0.1099 0.1083 0.0974 0.2280 0.2431 0.3771 0.5598 … V55 V56 V57 V58 V59 V60 Class 1 0.0072 0.0167 0.0180 0.0084 0.0090 0.0032 R 2 0.0094 0.0191 0.0140 0.0049 0.0052 0.0044 R 3 0.0180 0.0244 0.0316 0.0164 0.0095 0.0078 R
13. 13.

    • Feature Importance (Random Forest) Feature Importance library(caret) ca.fit <-

    train(Class ~ ., data = Sonar, method = "rf", ntree = 100) varImp(ca.fit) rf variable importance only 20 most important variables shown (out of 60) Overall V11 100.00 V48 96.25 V45 75.45 V13 71.74 V10 60.50 …
14. 14.

    • Feature Importance (Random Forest) Feature Importance plot(varImp(ca.fit), top =

    10)
15. 15.

    • caretͷvarImp͸ɺطఆͰख๏ґଘͷܭࢉํ๏Λར༻ • help(varImp) Feature Importance randomForest::importance(ca.fit$finalModel) %>% as.data.frame %>%

    tibble::rownames_to_column(var = 'variable') %>% arrange(desc(MeanDecreaseGini)) %>% head(20) varImp(ca.fit, scale = F) Overall V11 3.793 V48 3.684 V45 3.078 V13 2.970 V10 2.643 … variable MeanDecreaseGini 1 V11 3.792926 2 V48 3.683850 3 V45 3.077996 4 V13 2.970039 5 V10 2.642881 …
16. 16.

    Feature Importance • randomForest::importance Ͱ͸ҎԼͷ૊Έ߹ ΘͤͰಛ௃ྔͷॏཁ౓Λܭࢉ $MBTTJpDBUJPO 3FHSFTTJPO .FBO
    EFDSFBTF
    JO
    BDDVSBDZ
    DPNQVUFE
    GSPN
    QFSNVUJOH
    00#
    EBUB

    &SSPS
    3BUF
    .FBO%FDSFBTF"DDVSBDZ .4&
    *OD.4& .FBO
    EFDSFBTF
    JO
    OPEF
    JNQVSJUZ (JOJ
    JOEFY
    .FBO%FDSFBTF(JOJ 344
    *OD/PEF1VSJUZ
17. 17.

    OOB αϯϓϦϯά (෮ݩநग़) … OOB ࢀߟ: OOBσʔλ͔Βͷಛ௃ྔ ͷॏཁ౓ͷܭࢉ ֶश ֶश

    OOB ༧ଌ OOB ༧ଌ … ֤ྻͷ஋Λγϟοϑϧ͠ɺOOBޡΓ཰ͱൺֱ ༧ଌ ൺֱ … ༧ଌ ൺֱ
18. 18.

    • caret ͸Ϟσϧґଘ͠ͳ͍ॏཁ౓΋αϙʔτ Feature Importance (Model Agnostic) varImp(ca.fit, useModel =

    F) ROC curve variable importance only 20 most important variables shown (out of 60) Importance V11 100.00 V12 86.30 V10 82.64 V49 82.12 V9 81.97 …
19. 19.

    • Classification • આ໌ม਺͝ͱʹROC෼ੳ • ଟΫϥεͷ৔߹͸2Ϋϥεͷ૊Έ߹Θͤ͝ͱ • Regression • આ໌ม਺͝ͱʹઢܗճؼ/LOESS

    Feature Importance (Model Agnostic)
20. 20.

    ղऍՄೳੑ .PEFM
    4QFDJpD .PEFM
    "HOPTUJD (MPCBM
    *OUFSQSFUBCJMJUZ w 3FHSFTTJPO
    $PF⒏DJFOUT
    w 'FBUVSF
    *NQPSUBODF
    ʜ

    w 4VSSPHBUF
    .PEFMT
    w 4FOTJUJWJUZ
    "OBMZTJT
    ʜ -PDBM
    *OUFSQSFUBCJMJUZ w .BYJNVN
    "DUJWBUJPO
    "OBMZTJT
    ʜ w -*.&
    w -0$0
    w 4)"1
    ʜ
21. 21.

    Surrogate Models • ෳࡶͳϞσϧΛ୅ସ͢ΔγϯϓϧͳϞσϧΛ ௐ΂Δ Ideas on interpreting machine learning

    https://www.oreilly.com/ideas/ideas-on-interpreting-machine-learning Figure 14. An illustration of surrogate models for explaining a complex neural network. Figure courtesy of Patrick Hall and the H2O.ai team.
22. 22.

    Sensitivity Analysis • ͋Δಛ௃ྔΛมԽͤͨ࣌͞ͷϞσϧͷग़ྗΛ ௐ΂Δ Sensitivity analysis for neural networks

    https://beckmw.wordpress.com/2013/10/07/sensitivity-analysis-for-neural-networks/
23. 23.

    Partial Dependence • ͋Δಛ௃ྔΛมԽͤͨ࣌͞ͷϞσϧͷग़ྗΛ ௐ΂Δ library(pdp) library(ggplot2) pd <- partial(ca.fit,

    ɹɹpred.var = c("V11", “V9"), ɹchull = TRUE) autoplot(pd, contour = TRUE)
24. 24.

    ղऍՄೳੑ .PEFM
    4QFDJpD .PEFM
    "HOPTUJD (MPCBM
    *OUFSQSFUBCJMJUZ w 3FHSFTTJPO
    $PF⒏DJFOUT
    w 'FBUVSF
    *NQPSUBODF
    ʜ

    w 1BSUJBM
    %FQFOEFODF
    w 4VSSPHBUF
    .PEFMT
    w 4FOTJUJWJUZ
    "OBMZTJT
    ʜ -PDBM
    *OUFSQSFUBCJMJUZ w .BYJNVN
    "DUJWBUJPO
    "OBMZTJT
    ʜ w -*.&
    w -0$0
    w 4)"1
    ʜ
25. 25.

    LIMEͱ͸ʁ

26. 26.

    Local Interpretable Model-agnostic Explanations

27. 27.

    • ͋ΔσʔλͷۙลͰݩͷϞσϧΛ୅ସ͢ΔϞσϧΛ࡞੒ˠͦͷϞσϧͷ ಛ௃Λௐ΂Δ • Local Surrogate Model “Why Should I

    Trust You?” Explaining the Predictions of Any Classifier (Ribeiro, Singh, Guestrin 2016) LIME
28. 28.

    LIME • LIME͸ҎԼͷؔ਺Λ΋ͱʹσʔλ x ͷղऍΛಘΔ • G: ղऍ༻ͷֶशثͷू߹ • L:

    ղऍֶ͍ͨ͠शثͱղऍ༻ͷֶशثͷ ΠxͷݩͰͷࠩ • f: ղऍֶ͍ͨ͠शث • Πx: σʔλ x ͱͷྨࣅ౓ • Ω: ղऍ༻ͷֶशثͷෳࡶ͞ʹର͢Δേଇ߲ • ۩ମతखஈ͸υϝΠϯʹґଘ
29. 29.

    ςʔϒϧσʔλɾ෼ྨ໰୊ͷྫ • σʔλ x ͷपลͰαϯϓϦϯά • طఆͰ5,000 • αϯϓϦϯάํ๏͸ม਺ͷछྨʹґଘ •

    Exponential KernelͰॏΈ෇͚ • ม਺બ୒ • Forward/Backward, LARSͳͲ • RidgeճؼͳͲ ˞1ZUIPO࣮૷ ޙड़ ʹ΋ͱͮ͘
30. 30.

    ύοέʔδ • Python • ࿦จஶऀ࡞ • https://github.com/marcotcr/lime • R •

    ্هͷϙʔςΟϯά • https://github.com/thomasp85/lime install.packages(‘lime’)
31. 31.

    LIME (R) • αϯϓϧ library(caret) library(lime) model <- train(iris[-5], iris[[5]],

    method = 'rf') explainer <- lime(iris[-5], model) explanations <- explain(iris[1, -5], explainer, n_labels = 1, n_features = 2) explanations model_type case label label_prob model_r2 model_intercept 1 classification 1 setosa 1 0.3776584 0.2544468 2 classification 1 setosa 1 0.3776584 0.2544468 model_prediction feature feature_value feature_weight feature_desc 1 0.7113922 Sepal.Width 3.5 0.02101138 3.3 < Sepal.Width 2 0.7113922 Petal.Length 1.4 0.43593404 Petal.Length <= 1.60 data prediction 1 5.1, 3.5, 1.4, 0.2 1, 0, 0 2 5.1, 3.5, 1.4, 0.2 1, 0, 0 ֶशثΛ܇࿅ ղऍ༻ͷΫϥεΛ࡞੒ ղऍΛग़ྗ
32. 32.

    LIME (R) model_type case label label_prob model_r2 model_intercept 1 classification

    1 setosa 1 0.3776584 0.2544468 2 classification 1 setosa 1 0.3776584 0.2544468 model_prediction feature feature_value feature_weight feature_desc 1 0.7113922 Sepal.Width 3.5 0.02101138 3.3 < Sepal.Width 2 0.7113922 Petal.Length 1.4 0.43593404 Petal.Length <= 1.60 &YQMBOBUJPO NPEFM@UZQF 5IF
    UZQF
    PG
    UIF
    NPEFM
    VTFE
    GPS
    QSFEJDUJPO DBTF 5IF
    DBTF
    CFJOH
    FYQMBJOFE
    UIF
    SPXOBNF
    JO
    DBTFT  NPEFM@S 5IF
    RVBMJUZ
    PG
    UIF
    NPEFM
    VTFE
    GPS
    UIF
    FYQMBOBUJPO NPEFM@JOUFSDFQU 5IF
    JOUFSDFQU
    PG
    UIF
    NPEFM
    VTFE
    GPS
    UIF
    FYQMBOBUJPO NPEFM@QSFEJDUJPO 5IF
    QSFEJDUJPO
    PG
    UIF
    PCTFSWBUJPO
    CBTFE
    PO
    UIF
    NPEFM
    VTFE
    GPS
    UIF
    FYQMBOBUJPO GFBUVSF 5IF
    GFBUVSF
    VTFE
    GPS
    UIF
    FYQMBOBUJPO GFBUVSF@WBMVF 5IF
    WBMVF
    PG
    UIF
    GFBUVSF
    VTFE GFBUVSF@XFJHIU 5IF
    XFJHIU
    PG
    UIF
    GFBUVSF
    JO
    UIF
    FYQMBOBUJPO GFBUVSF@EFTD "
    IVNBO
    SFBEBCMF
    EFTDSJQUJPO
    PG
    UIF
    GFBUVSF
    JNQPSUBODF
33. 33.

    LIME (R) plot_features(explanations) ղऍΛϓϩοτ

34. 34.

    ςΩετσʔλͷྫ • ϥϯμϜʹ୯ޠΛআ֎ͨ͠σʔλΛੜ੒ • ίαΠϯྨࣅ౓ΛExponential KernelͰॏΈ෇͚ • Ҏ߱ɺςʔϒϧσʔλͱಉ༷ LIME -

    Local Interpretable Model-Agnostic Explanations https://homes.cs.washington.edu/~marcotcr/blog/lime/ ˞1ZUIPO࣮૷ʹ΋ͱͮ͘
35. 35.

    ը૾σʔλͷྫ Ideas on interpreting machine learning https://www.oreilly.com/ideas/ideas-on-interpreting-machine-learning Figure 15. An

    illustration of the LIME process in which a weighted linear model is used to explain a single prediction from a complex neural network. Figure courtesy of Marco Tulio Ribeiro; image used with permission. ˞1ZUIPO࣮૷ʹ΋ͱͮ͘ • ϥϯμϜʹηάϝϯτԽͨ͠σʔλΛੜ੒ • skimage.segmentationͷख๏͕ར༻Մ (quickshiftͳͲ) • ίαΠϯྨࣅ౓ͰॏΈ෇͚ • Ҏ߱ɺςʔϒϧσʔλͱಉ༷ ※RͰ͸ະ࣮૷
36. 36.

    ·ͱΊ .PEFM
    4QFDJpD .PEFM
    "HOPTUJD (MPCBM
    *OUFSQSFUBCJMJUZ w 3FHSFTTJPO
    $PF⒏DJFOUT
    w 'FBUVSF
    *NQPSUBODF
    ʜ

    w 4VSSPHBUF
    .PEFMT
    w 4FOTJUJWJUZ
    "OBMZTJT
    ʜ -PDBM
    *OUFSQSFUBCJMJUZ w .BYJNVN
    "DUJWBUJPO
    "OBMZTJT
    ʜ w -*.&
    w -0$0
    w 4)"1
    ʜ
37. 37.

    Enjoy!