SlideShare a Scribd company logo
1 of 32
Explaining the Explainability
‘Why’ and ‘How’ of Explainability
in Research
August 2021
Melih Bahar
➔ Data Scientist, Riskified
➔ Born in Turkey
➔ Loves whiskey!
melih.bahar@riskified.com
@melih_bhr
About Me
Riskified by the numbers
Global team,
nearly 50% in R&D
Countries across
the globe
Online volume
reviewed in 2020
650+ 180+
$60B+
50+
Publicly held companies
among our clients
98%
Client retention
rate in 2020
As of August 2021
Account Takeover (ATO)
A quick glance...
An ATO is when a bad actor gains access to another party’s legitimate account.
Agenda
01 Why
02 Explainability
03 SHAP
04 Research
Why?
Why now?
Explainable Accurate
Complex Model ✘ ✔
Simple Model ✔ ✘
Why
End Users POV
Loan?
Magic!
Model 55% Denied
Why?
Why
Data Scientist POV
“If you can’t explain it simply, you don’t understand it well enough.”
Albert Einstein
x1
.
.
xn
Data Model Interpretability
Methods
Humans
Validation
Performance
Trust, Informativeness,
Transferability, Fairness
Explainability
Explainability vs. Interpretability
Explainability
Explainability
An explainable model is a function that is too
complicated for a human to understand.
An additional method is needed to
understand how the model works.
Interpretability
A model is interpretable if it is capable
of being understood by humans on its
own.
Explainability
… is NOT causality!
Source: https://tuowang.rbind.io/project/causal-inference-notes/
SHAP Values
Shapley Values
A Short Introduction
The average of the marginal contributions across all permutations.
Source: https://towardsdatascience.com/explain-your-model-with-the-shap-values-bc36aac4de3d
Order matters!
Ann
Beth
Cindy
Beth
Ann Cindy
SHAP
SHapley Additive exPlanations
• Proposed by Lundberg and Lee (2016) - Based on Shapley Values
• A united approach to explaining the output of any machine learning model -
model agnostic
Properties
• Local Accuracy
• Consistency
• Missingness
Assumptions
• Independent Features
Global vs. Local
Global Interpretability
Provides explanations about the
general behavior of the model
over the entire population.
Local Interpretability
Provides explanations for a specified
prediction of the model
Explainable
Models in
Research
Model Comparison
• Not all models have out of the box feature importances (such as random forest
etc.)
- SHAP creates a common ground for comparison.
• Adds “explanation” to the “performance”
Tree-based Model (Boosting)
PRAUC: 0.829
Kernel-based Model (SVM)
PRAUC: 0.838
Model Debugging
Error Analysis on Instances
It is easier to detect that a selected feature should have a different effect than the observed one.
False Positive False Negative
Model Debugging
Error Analysis on Incorrect Predictions
We can take the subset of the errors of our models and see which features
affect the most for these incorrect classifications.
Model Debugging
Adding Domain Experts to the Loop
Feedback for/from analysts from/for the model helps getting more accurate
labels and find the logic flaws in the model.
ATO ? Legit ?
Feature Selection
• The wrapper-based methods (BORUTA, RFE etc.) may result in suboptimal performances.
• The standard feature importance method of decision trees tends to overestimate the importance
of continuous or high-cardinality categorical variables. SHAP reduces this bias.
Feature Selection
Effect of Specific Features
• Helps see how the values of a specific feature affect the predictions.
• Helps decide if a specific feature is helpful or not.
Feature Selection
Effect of Specific Features
Too high effect of a specific feature (with respect to the other features) may
indicate too high correlation to the label.
Feature Prioritization
Helps us decide which features (or group of features) are “more” important to develop in
production right now and which ones can wait.
For Production
Clustering
Creating the sub-segments
Clustering using the SHAP values can show that there are different sub-segments in our set.
Clustering
Understanding the Sub-segments
ATO type A ATO type B
And possibly more...
Disclaimers
The Catch
• Computationally expensive and time-consuming for large numbers of features.
• Explanations are generated too late in the machine learning pipeline.
• Provide no guarantee that your model will behave as expected in the future for new
data.
Available on MEDIUM
https://medium.com/riskified-technology
Melih Bahar
melih.bahar@riskified.com
@melih_bhr /Twitter
Thank You
For Your Time!

More Related Content

Similar to Explaining the Explainability: ‘Why’ and ‘How’ of Explainability in Research

Can AI finally "cure" the Marketing Myopia?
Can AI finally "cure" the Marketing Myopia?Can AI finally "cure" the Marketing Myopia?
Can AI finally "cure" the Marketing Myopia?Tathagat Varma
 
Responsible AI in Industry: Practical Challenges and Lessons Learned
Responsible AI in Industry: Practical Challenges and Lessons LearnedResponsible AI in Industry: Practical Challenges and Lessons Learned
Responsible AI in Industry: Practical Challenges and Lessons LearnedKrishnaram Kenthapadi
 
Practical Applications of Machine Learning in Cybersecurity
Practical Applications of Machine Learning in CybersecurityPractical Applications of Machine Learning in Cybersecurity
Practical Applications of Machine Learning in Cybersecurityscoopnewsgroup
 
GA – Client Project General Guidelines Mgmt5074 Fanshaw.docx
GA – Client Project General Guidelines Mgmt5074  Fanshaw.docxGA – Client Project General Guidelines Mgmt5074  Fanshaw.docx
GA – Client Project General Guidelines Mgmt5074 Fanshaw.docxhanneloremccaffery
 
201206 IASA Session 408 - Applied Analytics
201206 IASA Session 408 - Applied Analytics201206 IASA Session 408 - Applied Analytics
201206 IASA Session 408 - Applied AnalyticsSteven Callahan
 
Towards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptxTowards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptxLuis775803
 
Ethics in AI & Overcoming Product Bias by Smartsheet Sr Dir PM
Ethics in AI & Overcoming Product Bias by Smartsheet Sr Dir PMEthics in AI & Overcoming Product Bias by Smartsheet Sr Dir PM
Ethics in AI & Overcoming Product Bias by Smartsheet Sr Dir PMProduct School
 
Driving the Future of HR with Analytics and Bots
Driving the Future of HR with Analytics and Bots Driving the Future of HR with Analytics and Bots
Driving the Future of HR with Analytics and Bots Ahmad Areeb Faraz
 
WSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoT
WSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoTWSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoT
WSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoTWSO2
 
Presentation on 'Why cant people estimate' event, 23rd June 2016
Presentation on 'Why cant people estimate' event, 23rd June 2016 Presentation on 'Why cant people estimate' event, 23rd June 2016
Presentation on 'Why cant people estimate' event, 23rd June 2016 Association for Project Management
 
Predictive Analytics for Non-programmers
Predictive Analytics for Non-programmersPredictive Analytics for Non-programmers
Predictive Analytics for Non-programmersOlalekan Fuad Elesin
 
VS Live 2021 VST09 agile team metrics Fast Focus - angela dugan
VS Live 2021   VST09 agile team metrics Fast Focus - angela duganVS Live 2021   VST09 agile team metrics Fast Focus - angela dugan
VS Live 2021 VST09 agile team metrics Fast Focus - angela duganAngela Dugan
 
MonetizingStatistics
MonetizingStatisticsMonetizingStatistics
MonetizingStatisticsAaron Sankey
 
Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...
Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...
Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...Analytics India Magazine
 
Analytic Strategy Value Map
Analytic Strategy Value MapAnalytic Strategy Value Map
Analytic Strategy Value MapJaap Vink
 
Next Generation Compliance: Using Analytics to Reduce Compliance Risk
Next Generation Compliance: Using Analytics to Reduce Compliance RiskNext Generation Compliance: Using Analytics to Reduce Compliance Risk
Next Generation Compliance: Using Analytics to Reduce Compliance Riskqordata
 
Stan Gaidar: How far can you go improving user experience with AI?
Stan Gaidar: How far can you go improving user experience with AI? Stan Gaidar: How far can you go improving user experience with AI?
Stan Gaidar: How far can you go improving user experience with AI? Lviv Startup Club
 
Business Analyst Interview Questions with Answers
Business Analyst Interview Questions with AnswersBusiness Analyst Interview Questions with Answers
Business Analyst Interview Questions with AnswersMaria FutureThoughts
 

Similar to Explaining the Explainability: ‘Why’ and ‘How’ of Explainability in Research (20)

Can AI finally "cure" the Marketing Myopia?
Can AI finally "cure" the Marketing Myopia?Can AI finally "cure" the Marketing Myopia?
Can AI finally "cure" the Marketing Myopia?
 
Responsible AI in Industry: Practical Challenges and Lessons Learned
Responsible AI in Industry: Practical Challenges and Lessons LearnedResponsible AI in Industry: Practical Challenges and Lessons Learned
Responsible AI in Industry: Practical Challenges and Lessons Learned
 
Practical Applications of Machine Learning in Cybersecurity
Practical Applications of Machine Learning in CybersecurityPractical Applications of Machine Learning in Cybersecurity
Practical Applications of Machine Learning in Cybersecurity
 
GA – Client Project General Guidelines Mgmt5074 Fanshaw.docx
GA – Client Project General Guidelines Mgmt5074  Fanshaw.docxGA – Client Project General Guidelines Mgmt5074  Fanshaw.docx
GA – Client Project General Guidelines Mgmt5074 Fanshaw.docx
 
201206 IASA Session 408 - Applied Analytics
201206 IASA Session 408 - Applied Analytics201206 IASA Session 408 - Applied Analytics
201206 IASA Session 408 - Applied Analytics
 
Selling Text Analytics to your boss
Selling Text Analytics to your bossSelling Text Analytics to your boss
Selling Text Analytics to your boss
 
Towards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptxTowards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptx
 
Ethics in AI & Overcoming Product Bias by Smartsheet Sr Dir PM
Ethics in AI & Overcoming Product Bias by Smartsheet Sr Dir PMEthics in AI & Overcoming Product Bias by Smartsheet Sr Dir PM
Ethics in AI & Overcoming Product Bias by Smartsheet Sr Dir PM
 
Driving the Future of HR with Analytics and Bots
Driving the Future of HR with Analytics and Bots Driving the Future of HR with Analytics and Bots
Driving the Future of HR with Analytics and Bots
 
WSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoT
WSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoTWSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoT
WSO2Con USA 2015: Keynote - The Future of Real-Time Analytics and IoT
 
Presentation on 'Why cant people estimate' event, 23rd June 2016
Presentation on 'Why cant people estimate' event, 23rd June 2016 Presentation on 'Why cant people estimate' event, 23rd June 2016
Presentation on 'Why cant people estimate' event, 23rd June 2016
 
Predictive Analytics for Non-programmers
Predictive Analytics for Non-programmersPredictive Analytics for Non-programmers
Predictive Analytics for Non-programmers
 
VS Live 2021 VST09 agile team metrics Fast Focus - angela dugan
VS Live 2021   VST09 agile team metrics Fast Focus - angela duganVS Live 2021   VST09 agile team metrics Fast Focus - angela dugan
VS Live 2021 VST09 agile team metrics Fast Focus - angela dugan
 
MonetizingStatistics
MonetizingStatisticsMonetizingStatistics
MonetizingStatistics
 
Intro.pptx
Intro.pptxIntro.pptx
Intro.pptx
 
Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...
Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...
Deciphering AI - Unlocking the Black Box of AIML with State-of-the-Art Techno...
 
Analytic Strategy Value Map
Analytic Strategy Value MapAnalytic Strategy Value Map
Analytic Strategy Value Map
 
Next Generation Compliance: Using Analytics to Reduce Compliance Risk
Next Generation Compliance: Using Analytics to Reduce Compliance RiskNext Generation Compliance: Using Analytics to Reduce Compliance Risk
Next Generation Compliance: Using Analytics to Reduce Compliance Risk
 
Stan Gaidar: How far can you go improving user experience with AI?
Stan Gaidar: How far can you go improving user experience with AI? Stan Gaidar: How far can you go improving user experience with AI?
Stan Gaidar: How far can you go improving user experience with AI?
 
Business Analyst Interview Questions with Answers
Business Analyst Interview Questions with AnswersBusiness Analyst Interview Questions with Answers
Business Analyst Interview Questions with Answers
 

Recently uploaded

Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...
Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...
Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...Jack DiGiovanna
 
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...Sapana Sha
 
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Callshivangimorya083
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdfHuman37
 
GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]📊 Markus Baersch
 
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改atducpo
 
VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...
VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...
VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...Suhani Kapoor
 
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Serviceranjana rawat
 
Data Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptxData Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptxFurkanTasci3
 
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...Pooja Nehwal
 
Call Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceCall Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceSapana Sha
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 
From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...Florian Roscheck
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDRafezzaman
 
Call Girls In Mahipalpur O9654467111 Escorts Service
Call Girls In Mahipalpur O9654467111  Escorts ServiceCall Girls In Mahipalpur O9654467111  Escorts Service
Call Girls In Mahipalpur O9654467111 Escorts ServiceSapana Sha
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)jennyeacort
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsappssapnasaifi408
 
Customer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxCustomer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxEmmanuel Dauda
 

Recently uploaded (20)

Decoding Loan Approval: Predictive Modeling in Action
Decoding Loan Approval: Predictive Modeling in ActionDecoding Loan Approval: Predictive Modeling in Action
Decoding Loan Approval: Predictive Modeling in Action
 
E-Commerce Order PredictionShraddha Kamble.pptx
E-Commerce Order PredictionShraddha Kamble.pptxE-Commerce Order PredictionShraddha Kamble.pptx
E-Commerce Order PredictionShraddha Kamble.pptx
 
Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...
Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...
Building on a FAIRly Strong Foundation to Connect Academic Research to Transl...
 
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
Saket, (-DELHI )+91-9654467111-(=)CHEAP Call Girls in Escorts Service Saket C...
 
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
꧁❤ Greater Noida Call Girls Delhi ❤꧂ 9711199171 ☎️ Hard And Sexy Vip Call
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf
 
GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]
 
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
代办国外大学文凭《原版美国UCLA文凭证书》加州大学洛杉矶分校毕业证制作成绩单修改
 
VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...
VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...
VIP High Profile Call Girls Amravati Aarushi 8250192130 Independent Escort Se...
 
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
(PARI) Call Girls Wanowrie ( 7001035870 ) HI-Fi Pune Escorts Service
 
Data Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptxData Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptx
 
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...{Pooja:  9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
{Pooja: 9892124323 } Call Girl in Mumbai | Jas Kaur Rate 4500 Free Hotel Del...
 
Call Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceCall Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts Service
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 
From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
 
Call Girls In Mahipalpur O9654467111 Escorts Service
Call Girls In Mahipalpur O9654467111  Escorts ServiceCall Girls In Mahipalpur O9654467111  Escorts Service
Call Girls In Mahipalpur O9654467111 Escorts Service
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
 
Customer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxCustomer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptx
 

Explaining the Explainability: ‘Why’ and ‘How’ of Explainability in Research

Editor's Notes

  1. We do fraud prevention in ecommerce It can be at all stages of a transaction - during the login, at the checkout (order), after the order (abuse)
  2. Riskified tries to solve fraud from different sides, ATO (account takeover) is one of them and for this meetup we will mostly focus on that. 2 decisions - challenge/allow… ATO is 1, we are trying to detect ATOs, uses internal and external data. We built the best model, carefully engineered features, checked several algorithms, hyperparameter search and everything… model goes live, everything’s looking great… But suddenly we get messages from a specific merchant that we are challenging good logins (that they use to test). It can be that it’s ok but THEN it was test logins.
  3. The harder the question we are trying to solve, the more sophisticated the machine learning models tend to become, making it almost impossible to interpret. This might mean more features, complex algorithms or complex patterns. Should we build an accurate model or sacrifice on accuracy and build an interpretable model? Solution is to have a model simple enough that you can explain but also accurate enough to meet your needs.
  4. Story of John - he gets declined for a loan and asks why...
  5. Explainability enables tech developers to troubleshoot, debug and upgrade their models, as well as innovate new functionalities. Without explanations, if the model makes lots of bad predictions then it remains a mystery as to why.
  6. We’ll talk about 2 different terms but mostly they are used interchangeably. Same same but different… Where does ensemble get in to this graph - less interpretable! Keep it short!
  7. Explainability methods make transparent the correlations picked up by ML models… As a result, explaining the model with will not reveal causal effects. All predictive models implicitly assume that everyone will keep behaving the same way in the future, and therefore correlation patterns will stay constant. To understand what happens if someone starts behaving differently, we need to build causal models, which requires making assumptions and using the tools of causal analysis.
  8. I won’t get into theory, so theoretical questions i can answer or show you the relevant references.
  9. When applied to machine learning, we assume that each feature is a player in the game, all working together to maximize the prediction. 0 - the feature/player doesn’t contribute For machine learning ⇒ Coalition - all the options of all features (only 1, only 2.. None.. etc.)
  10. Local accuracy - The feature contributions must add up to the difference of prediction and the average of all predictions (baseline). Consistency - If a model changes so that the marginal contribution of a feature value increases or stays the same, the Shapley value also increases or stays the same Missingness - a missing feature gets an attribution of zero Independent features - without getting too much into theory and how it’s all actually calculated, if features are dependent, this may lead to putting too much weight on unlikely features because of the way shap values are calculated. With the packages, though, at least for Tree based models it’s taken care of.
  11. Shapley values for feature contributions do not directly come from a local regression model. In regression models, the coefficients represent the effect of a feature assuming all the other features are already in the model. It is well-known that the values of the regression coefficients highly depend on the collinearity of the feature of interest with the other features that are included in the model. To eliminate this bias, Shapley values calculate feature contributions by averaging across all permutations of the features joining the model.
  12. I won’t be getting into feature importance too much as Liran will be talking about it in a more detailed way. Notice that in kernel-based model there are no categorical features (gray) Different models with a similar performance can base their predictions on completely different relations extracted from the same data. Despite the differences, the explanations are very useful.
  13. Ca_14 for example was built to detect ATOs (the higher the value - the riskier)
  14. Just look at FP or FN to understand the main pattern. We can also look at them separately - only FP or only FN to catch more specific mistakes
  15. The model used for feature selection may differ (in parameter configuration or in the type) from the one used for final fitting and prediction. This may result in suboptimal performances. The standard methods tend to overestimate the importance of continuous or high-cardinality categorical variables. SHAP helps when we perform feature selection. Instead of using the default tree-based importance, we select the best features like the ones with the highest shapley values. It reduces the bias! We still do it recursively (just instead of impurity or gini index, we use shap values)
  16. Good for manual feature selection It’s called a dependence plot For example, here we can see that the same value (about 0) affects the prediction both highly positively and highly negatively.
  17. Worked hard, built 50 features but it takes time for DEV to implement it all in the production right away. You need to choose wisely as their capacity and time is limited so how do you decide? Here we can see the feature importance including the existing and new features.. You can also check the feature importance of only the new features... Only specific features? Or you want the whole group2 or 4?
  18. Some domain knowledge and prior information are needed for accurate number of clusters. The advantage of using shap values for clustering is that shap values for all features are on the same scale, and have the same unit. Not like regular features with different scales and that are harder to compare/compute distance. This also shows that the model sees those “segments” differently! it shouldn’t be done blindly and the scores/scales should be checked and taken care of if there is a need you just need to find the right way for your use case and for your values to get the most accurate of it.
  19. We can see different segments have completely different set of features that affect the predictions! The importance can be explained using different segment sets (instead of just using the existing trainset)
  20. Other possibilities I can think of but haven’t tried yet… Anomaly/Outlier Detection ?
  21. Exponentially increasing number of permutations Post-model analysis only The data might change/drift
  22. In theory, SHAP is the better approach as it provides mathematical guarantees for the accuracy and consistency of explanations.