SlideShare a Scribd company logo
1 of 40
Download to read offline
A Unified Approach
to Interpreting
Model Predictions
Rama Irsheidat
Scott M. Lundberg. et al.
Introduction SHAP
Additive Feature
Attribution Methods
Simple Properties
Uniquely Determine
Additive Feature
Attributions
02
01
04
03
TABLE OF CONTENTS
Computational and
User Study
Experiments
05
Conclusion
06
Why do we care so
much about
explainability in ML ?
Introduction
Example
Introduction
Example
Introduction
Understanding why a model makes a certain prediction can be as crucial as
the prediction’s accuracy in many applications
Introduction
The reason this problem exists is with in Large Datasets complex models tend
to be very accurate but hard to interpret
Introduction
Focus on explaining individual predictions one at a time.
Introduction
We are replacing the input to a summation that you would get in a linear
model with something that represents the importance of that feature in the
complicated model.
Interprets individual
model predictions based
on locally approximating
the model around a given
prediction.
Interprets the predictions
of deep networks.
Recursive prediction
explanation method for
deep learning.
Shapley regression
values
Quantitative
Input Influence
Shapley sampling
values
LIME
DeepLIFT
Layer-Wise
Relevance
Propagation
Feature importance for linear
models. This method requires
retraining the model on all
features. It assigns an
importance value to each
feature that represents the
effect on the model prediction
of including that feature.
Explaining any model by Applying sampling
approximations to Equation in Shapley reg. values ,
and Approximating the effect of removing a variable
from the model by integrating over samples from the
training dataset.
Proposes a sampling
approximation to Shapley
values that is nearly
identical to Shapley
sampling values.
Interprets individual
model predictions based
on locally approximating
the model around a given
prediction. Interprets the predictions
of deep networks.
Recursive prediction
explanation method for
deep learning.
Shapley regression
values
Quantitative
Input Influence
Shapley sampling
values
LIME
DeepLIFT
Layer-Wise
Relevance
Propagation
Feature importances for linear
models in the presence of
multicollinearity. This method
requires retraining the model
on all features. It assigns an
importance value to each
feature that represents the
effect on the model prediction
of including that feature.
Explain any model by Applying sampling
approximations to Equation in Shapley reg. values ,
and Approximating the effect of removing a variable
from the model by integrating over samples from the
training dataset.
Proposes a sampling
approximation to Shapley
values that is nearly
identical to Shapley
sampling values.
Additive Feature Attribution Methods
Additive Feature Attribution Methods
Interprets individual
model predictions based
on locally approximating
the model around a given
prediction. Interprets the predictions
of deep networks.
Recursive prediction
explanation method for
deep learning.
Shapley regression
values
Quantitative
Input Influence
Shapley sampling
values
LIME
DeepLIFT
Layer-Wise
Relevance
Propagation
Feature importances for linear
models in the presence of
multicollinearity. This method
requires retraining the model
on all features. It assigns an
importance value to each
feature that represents the
effect on the model prediction
of including that feature.
Explain any model by Applying sampling
approximations to Equation in Shapley reg. values ,
and Approximating the effect of removing a variable
from the model by integrating over samples from the
training dataset.
Proposes a sampling
approximation to Shapley
values that is nearly
identical to Shapley
sampling values.
Have some better
theoretical grounding
but slower
computation
Additive Feature Attribution Methods
Interprets individual
model predictions based
on locally approximating
the model around a given
prediction. Interprets the predictions
of deep networks.
Recursive prediction
explanation method for
deep learning.
Shapley regression
values
Quantitative
Input Influence
Shapley sampling
values
LIME
DeepLIFT
Layer-Wise
Relevance
Propagation
Feature importances for linear
models in the presence of
multicollinearity. This method
requires retraining the model
on all features. It assigns an
importance value to each
feature that represents the
effect on the model prediction
of including that feature.
Explain any model by Applying sampling
approximations to Equation in Shapley reg. values ,
and Approximating the effect of removing a variable
from the model by integrating over samples from the
training dataset.
Proposes a sampling
approximation to Shapley
values that is nearly
identical to Shapley
sampling values.
Have faster
estimation but less
guarantees
Interprets individual
model predictions based
on locally approximating
the model around a given
prediction.
Interprets the predictions
of deep networks.
Recursive prediction
explanation method for
deep learning.
Shapley regression
values
Quantitative
Input Influence
Shapley sampling
values
LIME
DeepLIFT
Layer-Wise
Relevance
Propagation
Feature importances for linear
models in the presence of
multicollinearity. This method
requires retraining the model
on all features. It assigns an
importance value to each
feature that represents the
effect on the model prediction
of including that feature.
Explain any model by Applying sampling
approximations to Equation in Shapley reg. values ,
and Approximating the effect of removing a variable
from the model by integrating over samples from the
training dataset.
Proposes a sampling
approximation to Shapley
values that is nearly
identical to Shapley
sampling values.
SHAP
How should we define importance of each
feature (φi (f , x))
Base rate of loan rejection or how often do people get their loans denied on
average?
Why am I 55 percent ?
We have to explain is this
35 percent difference
here .
So how can we do this?
We should just take the expected value of the output of our model (Base
rate), then we can just introduce a term into that conditional expectation.
Fact that John's 20, his
risk jumps up by 15
percent.
We should just take the expected value of the output of our model (Base
rate), then we can just introduce a term into that conditional expectation.
A very risky profession
and that jumps the risk
up to 70 percent.
We should just take the expected value of the output of our model (Base
rate), then we can just introduce a term into that conditional expectation.
We should just take the expected value of the output of our model (Base
rate), then we can just introduce a term into that conditional expectation.
He made a ton of money
in the stock market last
year. So his capital gains
pushes him down to 55
percent.
We have to explain is this
35 percent difference
here . So how can we do
this?
We've basically divided up how
we got from here to here by
conditioning one at a time on
all the features until we've
conditioned on all of them.
Example:
We can't just pick a particular order and think that we've solved it so what
do we do here?
Meaningless
Simple Properties Uniquely Determine Additive Feature Attributions
Means the output of the
explanation model matches
the original model for the
prediction being explained.
Requires features missing
in the original input to have
no impact.
If you change the original model such that a
feature has a larger impact in every possible
ordering, then that input's attribution
(importance) should not decrease.
Shapley Properties
Local accuracy Missingness
Consistency
SHAP values arise from averaging the φi values
across all possible orderings.
Very painful to compute.
Find an approximate solution
1. Model-Agnostic Approximations
1.1 Shapley sampling values
2.1 Kernel SHAP (Linear LIME + Shapley values)
Linear LIME (uses a linear explanation model) fit a linear model locally to the original
model that we're trying to explain.
Shapley values are the only possible solution that satisfies Properties 1-3 – local
accuracy, missingness and consistency.
This means we can now
estimate the Shapley
values using linear
regression.
2. Model-Specific Approximations
2.1 Linear SHAP
For linear models, SHAP values can be approximated directly from the model’s weight
coefficients.
2.2 Low-Order SHAP
3.2 Max SHAP
Calculating the probability that each input will increase the maximum value over every
other input.
2. Model-Specific Approximations
2.4 Deep SHAP (DeepLIFT + Shapley values)
DeepLIFT
Recursive prediction explanation method for deep learning that satisfies local accuracy
and missingness, we know that Shapley values represent the attribution values that
satisfy consistency.
Adapting DeepLIFT to become a compositional approximation of SHAP values, leading to
Deep SHAP.
Computational and User Study
Experiments
1. Computational Efficiency
Comparing Shapley sampling, SHAP, and LIME on both dense and sparse decision tree
models illustrates both the improved sample efficiency of Kernel SHAP and that values
from LIME can differ significantly from SHAP values that satisfy local accuracy and
consistency.
2. Consistency with Human Intuition
(A) Attributions of sickness score (B) Attributions of profit among three men
Participants were asked to assign importance for the output (the sickness score or
money won) among the inputs (i.e., symptoms or players). We found a much stronger
agreement between human explanations and SHAP than with other methods.
3. Explaining Class Differences
Explaining the output of a convolutional network trained on the MNIST digit dataset.
(A) Red areas increase the probability of that class, and blue areas decrease the
probability . Masked removes pixels in order to go from 8 to 3.
(B) The change in log odds when masking over 20 random images supports the use of
better estimates of SHAP values.
Conclusion
• The growing tension between the accuracy and interpretability of model
predictions has motivated the development of methods that help users
interpret predictions.
• The SHAP framework identifies the class of additive feature importance
methods (which includes six previous methods) and shows there is a
unique solution in this class that adheres to desirable properties.
• We presented several different estimation methods for SHAP values, along
with proofs and experiments showing that these values are desirable.
THANKS!

More Related Content

What's hot

Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine LearningSri Ambati
 
Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...
Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...
Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...Sri Ambati
 
Explainability and bias in AI
Explainability and bias in AIExplainability and bias in AI
Explainability and bias in AIBill Liu
 
Explainable AI in Healthcare
Explainable AI in HealthcareExplainable AI in Healthcare
Explainable AI in Healthcarevonaurum
 
Introduction to Machine Learning Classifiers
Introduction to Machine Learning ClassifiersIntroduction to Machine Learning Classifiers
Introduction to Machine Learning ClassifiersFunctional Imperative
 
Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...
Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...
Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...Sri Ambati
 
CounterFactual Explanations.pdf
CounterFactual Explanations.pdfCounterFactual Explanations.pdf
CounterFactual Explanations.pdfBong-Ho Lee
 
Notes of AI for everyone - by Andrew Ng
Notes of AI for everyone - by Andrew NgNotes of AI for everyone - by Andrew Ng
Notes of AI for everyone - by Andrew Ngmgopalani
 
General Tips for participating Kaggle Competitions
General Tips for participating Kaggle CompetitionsGeneral Tips for participating Kaggle Competitions
General Tips for participating Kaggle CompetitionsMark Peng
 
Machine Learning Explanations: LIME framework
Machine Learning Explanations: LIME framework Machine Learning Explanations: LIME framework
Machine Learning Explanations: LIME framework Deep Learning Italia
 
Explainable AI - making ML and DL models more interpretable
Explainable AI - making ML and DL models more interpretableExplainable AI - making ML and DL models more interpretable
Explainable AI - making ML and DL models more interpretableAditya Bhattacharya
 
Explainable AI (XAI) - A Perspective
Explainable AI (XAI) - A Perspective Explainable AI (XAI) - A Perspective
Explainable AI (XAI) - A Perspective Saurabh Kaushik
 
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...Sri Ambati
 
Supervised learning
Supervised learningSupervised learning
Supervised learningankit_ppt
 
Inverse Reinforcement Learning Algorithms
Inverse Reinforcement Learning AlgorithmsInverse Reinforcement Learning Algorithms
Inverse Reinforcement Learning AlgorithmsSungjoon Choi
 
Module 4: Model Selection and Evaluation
Module 4: Model Selection and EvaluationModule 4: Model Selection and Evaluation
Module 4: Model Selection and EvaluationSara Hooker
 
DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictionsAnton Kulesh
 

What's hot (20)

Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...
Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...
Explaining Black-Box Machine Learning Predictions - Sameer Singh, Assistant P...
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
 
Explainability and bias in AI
Explainability and bias in AIExplainability and bias in AI
Explainability and bias in AI
 
Explainable AI in Healthcare
Explainable AI in HealthcareExplainable AI in Healthcare
Explainable AI in Healthcare
 
eScience SHAP talk
eScience SHAP talkeScience SHAP talk
eScience SHAP talk
 
Introduction to Machine Learning Classifiers
Introduction to Machine Learning ClassifiersIntroduction to Machine Learning Classifiers
Introduction to Machine Learning Classifiers
 
Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...
Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...
Achieving Algorithmic Transparency with Shapley Additive Explanations (H2O Lo...
 
CounterFactual Explanations.pdf
CounterFactual Explanations.pdfCounterFactual Explanations.pdf
CounterFactual Explanations.pdf
 
Notes of AI for everyone - by Andrew Ng
Notes of AI for everyone - by Andrew NgNotes of AI for everyone - by Andrew Ng
Notes of AI for everyone - by Andrew Ng
 
General Tips for participating Kaggle Competitions
General Tips for participating Kaggle CompetitionsGeneral Tips for participating Kaggle Competitions
General Tips for participating Kaggle Competitions
 
Machine Learning Explanations: LIME framework
Machine Learning Explanations: LIME framework Machine Learning Explanations: LIME framework
Machine Learning Explanations: LIME framework
 
Explainable AI - making ML and DL models more interpretable
Explainable AI - making ML and DL models more interpretableExplainable AI - making ML and DL models more interpretable
Explainable AI - making ML and DL models more interpretable
 
Explainable AI (XAI) - A Perspective
Explainable AI (XAI) - A Perspective Explainable AI (XAI) - A Perspective
Explainable AI (XAI) - A Perspective
 
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
Scott Lundberg, Microsoft Research - Explainable Machine Learning with Shaple...
 
Supervised learning
Supervised learningSupervised learning
Supervised learning
 
Machine Learning
Machine LearningMachine Learning
Machine Learning
 
Inverse Reinforcement Learning Algorithms
Inverse Reinforcement Learning AlgorithmsInverse Reinforcement Learning Algorithms
Inverse Reinforcement Learning Algorithms
 
Module 4: Model Selection and Evaluation
Module 4: Model Selection and EvaluationModule 4: Model Selection and Evaluation
Module 4: Model Selection and Evaluation
 
DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictions
 

Similar to A Unified Approach to Interpreting Model Predictions (SHAP)

Understanding Black Box Models with Shapley Values
Understanding Black Box Models with Shapley ValuesUnderstanding Black Box Models with Shapley Values
Understanding Black Box Models with Shapley ValuesJonathan Bechtel
 
Interpretable ML
Interpretable MLInterpretable ML
Interpretable MLMayur Sand
 
Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Hayim Makabee
 
Machine learning Mind Map
Machine learning Mind MapMachine learning Mind Map
Machine learning Mind MapAshish Patel
 
Intepretable Machine Learning
Intepretable Machine LearningIntepretable Machine Learning
Intepretable Machine LearningAnkit Tewari
 
Machine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdfMachine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdfAijun Zhang
 
WIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual DiagnosticsWIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual DiagnosticsWomen in Analytics Conference
 
Steering Model Selection with Visual Diagnostics: Women in Analytics 2019
Steering Model Selection with Visual Diagnostics: Women in Analytics 2019Steering Model Selection with Visual Diagnostics: Women in Analytics 2019
Steering Model Selection with Visual Diagnostics: Women in Analytics 2019Rebecca Bilbro
 
GENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATION
GENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATIONGENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATION
GENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATIONijaia
 
Regularization_BY_MOHAMED_ESSAM.pptx
Regularization_BY_MOHAMED_ESSAM.pptxRegularization_BY_MOHAMED_ESSAM.pptx
Regularization_BY_MOHAMED_ESSAM.pptxMohamed Essam
 
Steering Model Selection with Visual Diagnostics
Steering Model Selection with Visual DiagnosticsSteering Model Selection with Visual Diagnostics
Steering Model Selection with Visual DiagnosticsMelissa Moody
 
MachineLlearning introduction
MachineLlearning introductionMachineLlearning introduction
MachineLlearning introductionThe IOT Academy
 
Machine Learning.pdf
Machine Learning.pdfMachine Learning.pdf
Machine Learning.pdfBeyaNasr1
 
Top 100+ Google Data Science Interview Questions.pdf
Top 100+ Google Data Science Interview Questions.pdfTop 100+ Google Data Science Interview Questions.pdf
Top 100+ Google Data Science Interview Questions.pdfDatacademy.ai
 
Overfitting & Underfitting
Overfitting & UnderfittingOverfitting & Underfitting
Overfitting & UnderfittingSOUMIT KAR
 
Sample_Subjective_Questions_Answers (1).pdf
Sample_Subjective_Questions_Answers (1).pdfSample_Subjective_Questions_Answers (1).pdf
Sample_Subjective_Questions_Answers (1).pdfAaryanArora10
 
Bengkel smartPLS 2011
Bengkel smartPLS 2011Bengkel smartPLS 2011
Bengkel smartPLS 2011Adi Ali
 
Loss Function.pptx
Loss Function.pptxLoss Function.pptx
Loss Function.pptxfunnyworld18
 

Similar to A Unified Approach to Interpreting Model Predictions (SHAP) (20)

Understanding Black Box Models with Shapley Values
Understanding Black Box Models with Shapley ValuesUnderstanding Black Box Models with Shapley Values
Understanding Black Box Models with Shapley Values
 
Interpretable ML
Interpretable MLInterpretable ML
Interpretable ML
 
Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)
 
Machine learning Mind Map
Machine learning Mind MapMachine learning Mind Map
Machine learning Mind Map
 
Intepretable Machine Learning
Intepretable Machine LearningIntepretable Machine Learning
Intepretable Machine Learning
 
Machine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdfMachine Learning Model Validation (Aijun Zhang 2024).pdf
Machine Learning Model Validation (Aijun Zhang 2024).pdf
 
WIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual DiagnosticsWIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual Diagnostics
 
Steering Model Selection with Visual Diagnostics: Women in Analytics 2019
Steering Model Selection with Visual Diagnostics: Women in Analytics 2019Steering Model Selection with Visual Diagnostics: Women in Analytics 2019
Steering Model Selection with Visual Diagnostics: Women in Analytics 2019
 
GENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATION
GENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATIONGENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATION
GENETIC ALGORITHM FOR FUNCTION APPROXIMATION: AN EXPERIMENTAL INVESTIGATION
 
Regularization_BY_MOHAMED_ESSAM.pptx
Regularization_BY_MOHAMED_ESSAM.pptxRegularization_BY_MOHAMED_ESSAM.pptx
Regularization_BY_MOHAMED_ESSAM.pptx
 
Steering Model Selection with Visual Diagnostics
Steering Model Selection with Visual DiagnosticsSteering Model Selection with Visual Diagnostics
Steering Model Selection with Visual Diagnostics
 
Introduction to ml
Introduction to mlIntroduction to ml
Introduction to ml
 
MachineLlearning introduction
MachineLlearning introductionMachineLlearning introduction
MachineLlearning introduction
 
Machine Learning.pdf
Machine Learning.pdfMachine Learning.pdf
Machine Learning.pdf
 
Top 100+ Google Data Science Interview Questions.pdf
Top 100+ Google Data Science Interview Questions.pdfTop 100+ Google Data Science Interview Questions.pdf
Top 100+ Google Data Science Interview Questions.pdf
 
Overfitting & Underfitting
Overfitting & UnderfittingOverfitting & Underfitting
Overfitting & Underfitting
 
Sample_Subjective_Questions_Answers (1).pdf
Sample_Subjective_Questions_Answers (1).pdfSample_Subjective_Questions_Answers (1).pdf
Sample_Subjective_Questions_Answers (1).pdf
 
Bengkel smartPLS 2011
Bengkel smartPLS 2011Bengkel smartPLS 2011
Bengkel smartPLS 2011
 
ML_in_QM_JC_02-10-18
ML_in_QM_JC_02-10-18ML_in_QM_JC_02-10-18
ML_in_QM_JC_02-10-18
 
Loss Function.pptx
Loss Function.pptxLoss Function.pptx
Loss Function.pptx
 

Recently uploaded

Data analytics courses in Nepal Presentation
Data analytics courses in Nepal PresentationData analytics courses in Nepal Presentation
Data analytics courses in Nepal Presentationanshikakulshreshtha11
 
AI Imagen for data-storytelling Infographics.pdf
AI Imagen for data-storytelling Infographics.pdfAI Imagen for data-storytelling Infographics.pdf
AI Imagen for data-storytelling Infographics.pdfMichaelSenkow
 
Exploratory Data Analysis - Dilip S.pptx
Exploratory Data Analysis - Dilip S.pptxExploratory Data Analysis - Dilip S.pptx
Exploratory Data Analysis - Dilip S.pptxDilipVasan
 
一比一原版阿德莱德大学毕业证成绩单如何办理
一比一原版阿德莱德大学毕业证成绩单如何办理一比一原版阿德莱德大学毕业证成绩单如何办理
一比一原版阿德莱德大学毕业证成绩单如何办理pyhepag
 
一比一原版西悉尼大学毕业证成绩单如何办理
一比一原版西悉尼大学毕业证成绩单如何办理一比一原版西悉尼大学毕业证成绩单如何办理
一比一原版西悉尼大学毕业证成绩单如何办理pyhepag
 
Fuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertaintyFuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertaintyRafigAliyev2
 
basics of data science with application areas.pdf
basics of data science with application areas.pdfbasics of data science with application areas.pdf
basics of data science with application areas.pdfvyankatesh1
 
Easy and simple project file on mp online
Easy and simple project file on mp onlineEasy and simple project file on mp online
Easy and simple project file on mp onlinebalibahu1313
 
2024 Q1 Tableau User Group Leader Quarterly Call
2024 Q1 Tableau User Group Leader Quarterly Call2024 Q1 Tableau User Group Leader Quarterly Call
2024 Q1 Tableau User Group Leader Quarterly Calllward7
 
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理pyhepag
 
2024 Q2 Orange County (CA) Tableau User Group Meeting
2024 Q2 Orange County (CA) Tableau User Group Meeting2024 Q2 Orange County (CA) Tableau User Group Meeting
2024 Q2 Orange County (CA) Tableau User Group MeetingAlison Pitt
 
Artificial_General_Intelligence__storm_gen_article.pdf
Artificial_General_Intelligence__storm_gen_article.pdfArtificial_General_Intelligence__storm_gen_article.pdf
Artificial_General_Intelligence__storm_gen_article.pdfscitechtalktv
 
一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理cyebo
 
Supply chain analytics to combat the effects of Ukraine-Russia-conflict
Supply chain analytics to combat the effects of Ukraine-Russia-conflictSupply chain analytics to combat the effects of Ukraine-Russia-conflict
Supply chain analytics to combat the effects of Ukraine-Russia-conflictJack Cole
 
How I opened a fake bank account and didn't go to prison
How I opened a fake bank account and didn't go to prisonHow I opened a fake bank account and didn't go to prison
How I opened a fake bank account and didn't go to prisonPayment Village
 
Generative AI for Trailblazers_ Unlock the Future of AI.pdf
Generative AI for Trailblazers_ Unlock the Future of AI.pdfGenerative AI for Trailblazers_ Unlock the Future of AI.pdf
Generative AI for Trailblazers_ Unlock the Future of AI.pdfEmmanuel Dauda
 
Pre-ProductionImproveddsfjgndflghtgg.pptx
Pre-ProductionImproveddsfjgndflghtgg.pptxPre-ProductionImproveddsfjgndflghtgg.pptx
Pre-ProductionImproveddsfjgndflghtgg.pptxStephen266013
 
一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理cyebo
 
Atlantic Grupa Case Study (Mintec Data AI)
Atlantic Grupa Case Study (Mintec Data AI)Atlantic Grupa Case Study (Mintec Data AI)
Atlantic Grupa Case Study (Mintec Data AI)Jon Hansen
 
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPsWebinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPsCEPTES Software Inc
 

Recently uploaded (20)

Data analytics courses in Nepal Presentation
Data analytics courses in Nepal PresentationData analytics courses in Nepal Presentation
Data analytics courses in Nepal Presentation
 
AI Imagen for data-storytelling Infographics.pdf
AI Imagen for data-storytelling Infographics.pdfAI Imagen for data-storytelling Infographics.pdf
AI Imagen for data-storytelling Infographics.pdf
 
Exploratory Data Analysis - Dilip S.pptx
Exploratory Data Analysis - Dilip S.pptxExploratory Data Analysis - Dilip S.pptx
Exploratory Data Analysis - Dilip S.pptx
 
一比一原版阿德莱德大学毕业证成绩单如何办理
一比一原版阿德莱德大学毕业证成绩单如何办理一比一原版阿德莱德大学毕业证成绩单如何办理
一比一原版阿德莱德大学毕业证成绩单如何办理
 
一比一原版西悉尼大学毕业证成绩单如何办理
一比一原版西悉尼大学毕业证成绩单如何办理一比一原版西悉尼大学毕业证成绩单如何办理
一比一原版西悉尼大学毕业证成绩单如何办理
 
Fuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertaintyFuzzy Sets decision making under information of uncertainty
Fuzzy Sets decision making under information of uncertainty
 
basics of data science with application areas.pdf
basics of data science with application areas.pdfbasics of data science with application areas.pdf
basics of data science with application areas.pdf
 
Easy and simple project file on mp online
Easy and simple project file on mp onlineEasy and simple project file on mp online
Easy and simple project file on mp online
 
2024 Q1 Tableau User Group Leader Quarterly Call
2024 Q1 Tableau User Group Leader Quarterly Call2024 Q1 Tableau User Group Leader Quarterly Call
2024 Q1 Tableau User Group Leader Quarterly Call
 
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
一比一原版(Monash毕业证书)莫纳什大学毕业证成绩单如何办理
 
2024 Q2 Orange County (CA) Tableau User Group Meeting
2024 Q2 Orange County (CA) Tableau User Group Meeting2024 Q2 Orange County (CA) Tableau User Group Meeting
2024 Q2 Orange County (CA) Tableau User Group Meeting
 
Artificial_General_Intelligence__storm_gen_article.pdf
Artificial_General_Intelligence__storm_gen_article.pdfArtificial_General_Intelligence__storm_gen_article.pdf
Artificial_General_Intelligence__storm_gen_article.pdf
 
一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理一比一原版纽卡斯尔大学毕业证成绩单如何办理
一比一原版纽卡斯尔大学毕业证成绩单如何办理
 
Supply chain analytics to combat the effects of Ukraine-Russia-conflict
Supply chain analytics to combat the effects of Ukraine-Russia-conflictSupply chain analytics to combat the effects of Ukraine-Russia-conflict
Supply chain analytics to combat the effects of Ukraine-Russia-conflict
 
How I opened a fake bank account and didn't go to prison
How I opened a fake bank account and didn't go to prisonHow I opened a fake bank account and didn't go to prison
How I opened a fake bank account and didn't go to prison
 
Generative AI for Trailblazers_ Unlock the Future of AI.pdf
Generative AI for Trailblazers_ Unlock the Future of AI.pdfGenerative AI for Trailblazers_ Unlock the Future of AI.pdf
Generative AI for Trailblazers_ Unlock the Future of AI.pdf
 
Pre-ProductionImproveddsfjgndflghtgg.pptx
Pre-ProductionImproveddsfjgndflghtgg.pptxPre-ProductionImproveddsfjgndflghtgg.pptx
Pre-ProductionImproveddsfjgndflghtgg.pptx
 
一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理一比一原版麦考瑞大学毕业证成绩单如何办理
一比一原版麦考瑞大学毕业证成绩单如何办理
 
Atlantic Grupa Case Study (Mintec Data AI)
Atlantic Grupa Case Study (Mintec Data AI)Atlantic Grupa Case Study (Mintec Data AI)
Atlantic Grupa Case Study (Mintec Data AI)
 
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPsWebinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
Webinar One View, Multiple Systems No-Code Integration of Salesforce and ERPs
 

A Unified Approach to Interpreting Model Predictions (SHAP)

  • 1. A Unified Approach to Interpreting Model Predictions Rama Irsheidat Scott M. Lundberg. et al.
  • 2. Introduction SHAP Additive Feature Attribution Methods Simple Properties Uniquely Determine Additive Feature Attributions 02 01 04 03 TABLE OF CONTENTS Computational and User Study Experiments 05 Conclusion 06
  • 3. Why do we care so much about explainability in ML ?
  • 6. Introduction Understanding why a model makes a certain prediction can be as crucial as the prediction’s accuracy in many applications
  • 7. Introduction The reason this problem exists is with in Large Datasets complex models tend to be very accurate but hard to interpret
  • 8. Introduction Focus on explaining individual predictions one at a time.
  • 9. Introduction We are replacing the input to a summation that you would get in a linear model with something that represents the importance of that feature in the complicated model.
  • 10. Interprets individual model predictions based on locally approximating the model around a given prediction. Interprets the predictions of deep networks. Recursive prediction explanation method for deep learning. Shapley regression values Quantitative Input Influence Shapley sampling values LIME DeepLIFT Layer-Wise Relevance Propagation Feature importance for linear models. This method requires retraining the model on all features. It assigns an importance value to each feature that represents the effect on the model prediction of including that feature. Explaining any model by Applying sampling approximations to Equation in Shapley reg. values , and Approximating the effect of removing a variable from the model by integrating over samples from the training dataset. Proposes a sampling approximation to Shapley values that is nearly identical to Shapley sampling values.
  • 11. Interprets individual model predictions based on locally approximating the model around a given prediction. Interprets the predictions of deep networks. Recursive prediction explanation method for deep learning. Shapley regression values Quantitative Input Influence Shapley sampling values LIME DeepLIFT Layer-Wise Relevance Propagation Feature importances for linear models in the presence of multicollinearity. This method requires retraining the model on all features. It assigns an importance value to each feature that represents the effect on the model prediction of including that feature. Explain any model by Applying sampling approximations to Equation in Shapley reg. values , and Approximating the effect of removing a variable from the model by integrating over samples from the training dataset. Proposes a sampling approximation to Shapley values that is nearly identical to Shapley sampling values. Additive Feature Attribution Methods
  • 12. Additive Feature Attribution Methods Interprets individual model predictions based on locally approximating the model around a given prediction. Interprets the predictions of deep networks. Recursive prediction explanation method for deep learning. Shapley regression values Quantitative Input Influence Shapley sampling values LIME DeepLIFT Layer-Wise Relevance Propagation Feature importances for linear models in the presence of multicollinearity. This method requires retraining the model on all features. It assigns an importance value to each feature that represents the effect on the model prediction of including that feature. Explain any model by Applying sampling approximations to Equation in Shapley reg. values , and Approximating the effect of removing a variable from the model by integrating over samples from the training dataset. Proposes a sampling approximation to Shapley values that is nearly identical to Shapley sampling values. Have some better theoretical grounding but slower computation
  • 13. Additive Feature Attribution Methods Interprets individual model predictions based on locally approximating the model around a given prediction. Interprets the predictions of deep networks. Recursive prediction explanation method for deep learning. Shapley regression values Quantitative Input Influence Shapley sampling values LIME DeepLIFT Layer-Wise Relevance Propagation Feature importances for linear models in the presence of multicollinearity. This method requires retraining the model on all features. It assigns an importance value to each feature that represents the effect on the model prediction of including that feature. Explain any model by Applying sampling approximations to Equation in Shapley reg. values , and Approximating the effect of removing a variable from the model by integrating over samples from the training dataset. Proposes a sampling approximation to Shapley values that is nearly identical to Shapley sampling values. Have faster estimation but less guarantees
  • 14. Interprets individual model predictions based on locally approximating the model around a given prediction. Interprets the predictions of deep networks. Recursive prediction explanation method for deep learning. Shapley regression values Quantitative Input Influence Shapley sampling values LIME DeepLIFT Layer-Wise Relevance Propagation Feature importances for linear models in the presence of multicollinearity. This method requires retraining the model on all features. It assigns an importance value to each feature that represents the effect on the model prediction of including that feature. Explain any model by Applying sampling approximations to Equation in Shapley reg. values , and Approximating the effect of removing a variable from the model by integrating over samples from the training dataset. Proposes a sampling approximation to Shapley values that is nearly identical to Shapley sampling values.
  • 15. SHAP
  • 16. How should we define importance of each feature (φi (f , x))
  • 17. Base rate of loan rejection or how often do people get their loans denied on average?
  • 18. Why am I 55 percent ?
  • 19. We have to explain is this 35 percent difference here . So how can we do this?
  • 20. We should just take the expected value of the output of our model (Base rate), then we can just introduce a term into that conditional expectation. Fact that John's 20, his risk jumps up by 15 percent.
  • 21. We should just take the expected value of the output of our model (Base rate), then we can just introduce a term into that conditional expectation. A very risky profession and that jumps the risk up to 70 percent.
  • 22. We should just take the expected value of the output of our model (Base rate), then we can just introduce a term into that conditional expectation.
  • 23. We should just take the expected value of the output of our model (Base rate), then we can just introduce a term into that conditional expectation. He made a ton of money in the stock market last year. So his capital gains pushes him down to 55 percent.
  • 24. We have to explain is this 35 percent difference here . So how can we do this? We've basically divided up how we got from here to here by conditioning one at a time on all the features until we've conditioned on all of them.
  • 25.
  • 27. We can't just pick a particular order and think that we've solved it so what do we do here? Meaningless
  • 28. Simple Properties Uniquely Determine Additive Feature Attributions
  • 29. Means the output of the explanation model matches the original model for the prediction being explained. Requires features missing in the original input to have no impact. If you change the original model such that a feature has a larger impact in every possible ordering, then that input's attribution (importance) should not decrease. Shapley Properties Local accuracy Missingness Consistency
  • 30. SHAP values arise from averaging the φi values across all possible orderings. Very painful to compute.
  • 32. 1. Model-Agnostic Approximations 1.1 Shapley sampling values 2.1 Kernel SHAP (Linear LIME + Shapley values) Linear LIME (uses a linear explanation model) fit a linear model locally to the original model that we're trying to explain. Shapley values are the only possible solution that satisfies Properties 1-3 – local accuracy, missingness and consistency. This means we can now estimate the Shapley values using linear regression.
  • 33. 2. Model-Specific Approximations 2.1 Linear SHAP For linear models, SHAP values can be approximated directly from the model’s weight coefficients. 2.2 Low-Order SHAP 3.2 Max SHAP Calculating the probability that each input will increase the maximum value over every other input.
  • 34. 2. Model-Specific Approximations 2.4 Deep SHAP (DeepLIFT + Shapley values) DeepLIFT Recursive prediction explanation method for deep learning that satisfies local accuracy and missingness, we know that Shapley values represent the attribution values that satisfy consistency. Adapting DeepLIFT to become a compositional approximation of SHAP values, leading to Deep SHAP.
  • 35. Computational and User Study Experiments
  • 36. 1. Computational Efficiency Comparing Shapley sampling, SHAP, and LIME on both dense and sparse decision tree models illustrates both the improved sample efficiency of Kernel SHAP and that values from LIME can differ significantly from SHAP values that satisfy local accuracy and consistency.
  • 37. 2. Consistency with Human Intuition (A) Attributions of sickness score (B) Attributions of profit among three men Participants were asked to assign importance for the output (the sickness score or money won) among the inputs (i.e., symptoms or players). We found a much stronger agreement between human explanations and SHAP than with other methods.
  • 38. 3. Explaining Class Differences Explaining the output of a convolutional network trained on the MNIST digit dataset. (A) Red areas increase the probability of that class, and blue areas decrease the probability . Masked removes pixels in order to go from 8 to 3. (B) The change in log odds when masking over 20 random images supports the use of better estimates of SHAP values.
  • 39. Conclusion • The growing tension between the accuracy and interpretability of model predictions has motivated the development of methods that help users interpret predictions. • The SHAP framework identifies the class of additive feature importance methods (which includes six previous methods) and shows there is a unique solution in this class that adheres to desirable properties. • We presented several different estimation methods for SHAP values, along with proofs and experiments showing that these values are desirable.