Automated Machine Learning and eXplainable Artificial Intelligence are disruptive technologies in Data Science. Here I briefly introduce them and show how DALEXverse may be used in better model development.

1. XAI or DIE
Przemysław Biecek
Samsung SRPOL
Warsaw University of Technology
Data Science Summit 2019

2. AutoML
eXplainable AI

3. AutoML

5. Auto-Keras
auto-sklearn
automl-gs
Auto-Weka
FeatureTools
h2o automl
Ludwig
mljar-supervised
Neural Network
Intelligence (NNI)
tpot
TransmografAI
Auto_ml

6. https://github.com/hibayesian/awesome-automl-papers

8. Defaults (package
defaults (Def.P) and
optimal defaults
(Def.O)),
tunability of the
hyperparameters
with the package
defaults (Tun.P) and
our optimal defaults
(Tun.O) as reference
and tuning space
quantiles (q0.05 and
q0.95) for different
parameters of the
algorithms

10. https://www.linkedin.com/in/igor2k/

11. eXplainable AI

13. https://www.wired.com/2015/10/can-learn-epic-failure-google-flu-trends/

14. https://www.massdevice.com/report-ibm-watson-delivered-unsafe-and-inaccurate-
cancer-recommendations/

15. https://www.propublica.org/article/how-we-analyzed-the-compas-recidivism-algorithm

17. • “You don’t see a lot of skepticism,” she says. “The algorithms are like shiny new
toys that we can’t resist using. We trust them so much that we project meaning on to
them.”
• Ultimately algorithms, according to O’Neil, reinforce discrimination and widen
inequality, “using people’s fear and trust of mathematics to prevent them from
asking questions”.
https://www.theguardian.com/books/2016/oct/27/cathy-oneil-weapons-of-math-
destruction-algorithms-big-data !17
Cathy O'Neil:
The era of blind faith
in big data must end
black boxes
Why do we need explanations for complex models?

18. Right to explanation
!18
Why do we need explanations for complex models?

19. https://panoptykon.org/wiadomosc/prawo-do-wyjasnienia-decyzji-kredytowej-dla-
kazdego-sukces-panoptykonu

22. !22
Local Model approximations
"Why Should I Trust You?" Explaining the Predictions of Any Classifier.
Marco Tulio Ribeiro, Sameer Singh, Carlos Guestrin (2016). https://arxiv.org/pdf/1602.04938.pdf
Port to R: Thomas Lin Pedersen (2017) https://github.com/thomasp85/lime
Other implementations: live (Staniak, Biecek 2018) and iml (Molnar 2018)
A different approach to model explanation is to locally approximate
the complex black-box model with an easier to interpret white-box
model constructed on interpretable features.

23. LIME / live
vs
Break Down
Model debugging

24. Biecek P (2018). “DALEX: Explainers for Complex Predictive Models in R.”
Journal of Machine Learning Research, 19(84), 1-5. URL:http://jmlr.org/papers/v19/18-416.html>

25. Chat bot dla wyjaśnień
https://kmichael08.github.io

26. Chat bot dla wyjaśnień
https://kmichael08.github.io
What If?
Why?

28. https://www.encyclopedia-titanica.org/
What are the odds of surviving?

29. What If?

30. Ceteris Paribus
Individual Conditional Expectations

31. Champion - Challenger

32. Why?

33. iBreakDown: Uncertainty of Model Explanations for Non-additive Predictive Models
Alicja Gosiewska, Przemyslaw Biecek (2019) https://arxiv.org/abs/1903.11420v1

34. SHAP (SHapley Additive exPlanations) Lundberg (2017)
IME complexity is O(2
p
). Shapley values are known for some
time and we have methods to approximate them efficiently.

35. Next
Generation
Explainers

36. What If? Comparison of models between languages

37. https://breakdeeper.netlify.com/

38. https://github.com/MI2DataLab/modelDown

39. https://chudekm.shinyapps.io/model_explorer_example/

40. What If? Interactive with D3

41. Predictive Models: Visual Exploration, Explanation and Debugging
Production
Development
Concept ValidateForge
Model debugging
Model development is an
iterative process. Each
iteration brings new insights.
Early phases:
Crisp modeling, general
understanding of the problem.
Medium phases:
Selective modeling, here we
select the best type of model.
Late phases:
Fine tuning of model
parameters or variable
engineering .
In each iteration model development starts with some concepts, ideas, then the model is trained and finally
model needs to be validated.
Predictions need to be
explained.
Here the instance level
explanation helps.
With time the model
performance may deteriorate,
thus it requires constant
monitoring, e.g. with the drifter
package.
Drop-out loss
Variable importance
GBM
baseline
ditrict
surface
floor
construction.year
no. rooms
full model
250 500 750 1000 1250
3000
3200
3400
surface
prediction
3600
3800
Surface
6019 100 148
Random Forest
feature influence
Variables attributions
GBM
intercept
district: Srodmiescie
surface: 22
no.rooms: 2
construction.year: 2005
floor: 1
prediction
Random Forest
intercept
district: Srodmiescie
surface: 22
no.rooms: 2
construction.year: 2005
floor: 1
prediction
LM
intercept
district: Srodmiescie
surface: 22
no.rooms: 2
construction.year: 2005
floor: 1
prediction
2000 2100 2200 2300 2400 2500 2600 2700 2800 2900
2046
2614.9
+358
+160
+78
-39.5
+12.4
2800
2425
-338
-112
+74
-53
+26
2378
2324.9
-239
+160
+68
+39.5
-12.4
feature influence
Variables attributions
GBM
intercept
district: Srodmiescie
surface: 22
no.rooms: 2
construction.year: 2005
floor: 1
prediction
2000 2100 2200 2300 2400 2500 2600 2700
2046
2614.9
+358
+160
+78
-39.5
+12.4
3000
3200
3400
surface
prediction
3600
3800
Surface
6019 100 148
Random Forest
3000
3200
3400
2 4 6
prediction
3600
3800
3000
3200
3400
3600
3800
Surface
1920 1940 1960 1980 2010
no.roomsfloor
surfaceconstruction.year
4020 80 120 144
2.50.8 1.15.0 7.5 11.2
Variable selection
Feature engineering
Random Forest
Factor Merger
Srodmiescie
Ochota
Mokotow
Zoliborz
Ursus
Bielany
Bemowo
Wola
Ursynow
Praga
0 2000 4000
group frequencyname price mean
5109.19
3954.83
3946.96
3918.55
3058.52
3045.79
3028.58
3011.69
3009.72
2991.48
nr
1
2
2
3
4
4
5
6
6
7
Prediction explanations What-If analysis Concept drift detection
explain

42. Liked it? Let us know! Found a bug? Fire an issue!