LIME

@Tokyo.R 66
https://atnd.org/events/92993

Transcript

1. LIME
   Masaaki Horikoshi @ ARISE analytics
   

   View Slide

2. ࣗݾ঺հ
   • R
   • ύοέʔδ։ൃͳͲ
   • Git Awards ࠃ಺1Ґ
   • Python
   • http://git-awards.com/users/search?login=sinhrks
   

   View Slide

3. Α͋͘Δ͜ͱ
   ΤʔΞΠͰ͍͍ײ͡ʹ΍ͬͱ͍ͯΑʂ
   ݁Ռ͕ྑ͚Ε͹த਎͸ؾʹ͠ͳ͍Αʂʂ
   Ͱɺ͜Εͬͯ݁ہͲ͏͍͏͜ͱͳͷʁ
   த਎͕Θ͔Βͳ͍΋ͷ͸࢖͑ͳ͍Αʂʂʂ
   ͑Β͍ਓ
   ˞΁ʔγϟͰ͸ͳ͍
   ݁Ռ͕ग़Δͱʜ
   

   View Slide

4. Interpretability
   ղऍՄೳੑ
   

   View Slide

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

   View Slide

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

   View Slide

7. ղऍՄೳੑ
   • ద੾ͳख๏͸ʮԿΛʯղऍ͍͔ͨ͠ʹґଘ
   .PEFM4QFDJpD .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
   

   View Slide

8. LIMEͱ͸ʁ
   

   View Slide

9. Local Interpretable
   Model-agnostic Explanations
   

   View Slide

10. LIME
    • “Why Should I Trust You?” Explaining the Predictions of Any Classifier
    (2016)
    • Marco Tulio Ribeiro, Sameer Singh, Carlos Guestrin
    

    View Slide

11. 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
    

    View Slide

12. ςʔϒϧσʔλɾ෼ྨ໰୊ͷྫ
    • σʔλ x ͷपลͰαϯϓϦϯά
    • طఆͰ5,000
    • αϯϓϦϯάํ๏͸ม਺ͷछྨʹґଘ
    • Exponential KernelͰॏΈ෇͚
    • ม਺બ୒
    • Forward/Backward, LARSͳͲ
    • RidgeճؼͳͲ
    ˞1ZUIPO࣮૷ ޙड़
    ʹ΋ͱͮ͘
    

    View Slide

13. ύοέʔδ
    • Python
    • ࿦จஶऀ࡞
    • https://github.com/marcotcr/lime
    • R
    • ্هͷϙʔςΟϯά
    • https://github.com/thomasp85/lime
    install.packages(‘lime’)
    

    View Slide

14. 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
    

    View Slide

15. LIME (R)
    plot_features(explanations)
    ղऍΛϓϩοτ
    

    View Slide

16. Enjoy!
    

    View Slide