SlideShare una empresa de Scribd logo
1 de 55
Descargar para leer sin conexión
Explainability and Bias in
ML/AI Models
Naveen Sundar Govindarajulu
August 9, 2019
Visit
and
sign up
RealityEngines.AI
Why now?
Life
Impacting
ML & AI
models
COMPAS
https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencingFrom:
Non recidivating black people twice as likely to be labelled high risk than non
recidivating white people
Why Explainability?
• More use of ML/AI models by laypersons.
• Laypersons need explanations
• Developers also need quick explanations to debug models
faster
• There may be a legal need for explanations:
• If you deny someone a loan, you may need to explain the
reason for the denial.
Explainability
Explainability using Interpretable Models
Prior offenses <= 0
Low Risk High Risk
Armed offense?
Med Risk
YES
NO
NO YES
Explainability vs Performance
Tradeoff
• Some machine learning models are more explainable than
others.
Performance
Explainability
Deep learning models
Linear Models
DecisionTrees
Explainability Method:
Feature Attribution
Classifier
Explainer
features
“Weights” for
features
Input
features
Output
What Features?
Interpretable Features
• We need interpretable features.
• Difficult for laypersons to understand raw feature spaces (e.g.
word embeddings)
• Humans are good at understanding presence or absence of
components.
Interpretable Instance
• E.g.
• For Text:
• Convert to a binary vector indicating presence or absence
of words
• For images
• Convert to a binary vector indicating presence or absence
of pixels or contiguous regions.
Method 1: LIME
From
https://github.com/marcotcr/lime
Locally Interpretable Model-agnostic
Explanations
Ribeiro, M.T., Singh, S. and Guestrin, C., 2016, August. Why Should I Trust
You?: Explaining the Predictions of Any Classifier. In Proceedings of the 22nd
ACM SIGKDD International Conference on Knowledge Discovery and Data
Mining (pp. 1135-1144). ACM.
Method 1: LIME
Any classifier
1 1 0 1 1 0 1 0 0 1 0
0 0 0 1 0 1 1 1 1 0 1
-2.1 1.1 -0.5 2.2 -1.2 -1.5 1 -3 0.8 5.6 1.5
Weights for the linear classifier then
give us feature importances
Binary vectors
-2.1 2.2 -3 5.6
Enforce
sparsity
Example:
Text Sentiment Classification
“The movie is not bad”
This movie is not bad
0 0 0 2.3 -1.5
Explanation for “Cat”
LIME with Images
From
https://github.com/marcotcr/lime
Ribeiro, M.T., Singh, S. and Guestrin, C., 2016, August. Why Should I Trust You?: Explaining the Predictions of Any Classifier. In
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135-1144). ACM.
Explanations for Multi-Label
Classifiers
Ribeiro, M.T., Singh, S. and Guestrin, C., 2016, August. Why Should I Trust You?: Explaining the Predictions of Any Classifier. In
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135-1144). ACM.
Using LIME for Debugging (E.g. 1)
https://github.com/marcotcr/lime
Using LIME for Debugging (E.g. 2)
https://github.com/marcotcr/lime
Using LIME for Debugging (E.g. 2)
Method 2: SHAP
Unifies many different feature attribution methods and has some
desirable properties.
1. LIME
2. Integrated Gradients
3. Shapley values
4. DeepLift
Lundberg, S.M. and Lee, S.I., 2017. A unified approach to interpreting model predictions. In
Advances in Neural Information Processing Systems (pp. 4765-4774).
Method 2: SHAP
• Derives from game-theoretic foundations.
• Shapley values used in game theory to assign values to players
in cooperative games.
What are Shapley values?
• Suppose there is a set S of N players
participating in a game with payoff for any S
subset of players participating in the game
given by:
• Shapley values provide one fair
way of dividing up the total
payoff among the N players.
ShapleyValue
Payoff for the group
including player i
Shapley value for player i
Payoff for a group without player i
SHAP Explanations
• Players are features.
• Payoff is the model’s real valued prediction.
SHAP Implementation
(https://github.com/slundberg/shap)
Different kinds of explainers:
1. TreeExplainer: fast and exact SHAP values for tree ensembles
2. KernelExplainer: approximate explainer for black box estimators
3. DeepExplainer: high-speed approximate explainer for deep learning models.
4. ExpectedGradients: SHAP-based extension of integrated gradients
XGBoost on UCI Income Dataset
Output is probability of income
over 50k
f87
f23
f23 f3
f34
f41
Base ValueOutput
Note: SHAP values are Model
Dependent.
Model 1
Model 2
Is This Form of Explainability
Enough?
• Explainability does not provide us with recourse.
• Recourse: Information needed to change a specific prediction to a
desired value.
• “If you had paid your credit card balance in full for the last three
months, you would have got that loan.”
Issues with SHAP and LIME
SHAP and LIME values are highly variable for instances that are very similar for
non-linear models.

On the Robustness of Interpretability Methods
https://arxiv.org/abs/1806.08049
Issues with SHAP and LIME
SHAP and LIME values are highly variable for instances that are very similar for
non-linear models.

On the Robustness of Interpretability Methods
https://arxiv.org/abs/1806.08049
Issues with SHAP and LIME
SHAP and LIME values don’t provide insight into how the model will behave on new instances.
https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16982
High-Precision Model-Agnostic Explanations
Take-home message
• Explainability is possible need not come at the cost of
performance.
• Explainability is not enough
• Recourse, etc.
Bias
Fairness and Bias in Machine
Learning
1. Bias in this context is unfairness (more or less).
2. Note we are not talking about standard statistical bias in machine
learning (the bias in the bias vs. variance tradeoff).
3. For completeness, this is one definition of statistical bias in machine
learning.
• Bias = Expected value of model - true value
Definitions of Fairness or Bias
1. Many, many, many definitions exists.
2. Application dependent. No one definition is better.
3. See “21 Definitions of Fairness” tutorial by Arvind Narayanan,ACM
2018 FAT*.
1. Key Point: Dozens of definitions exist (and not just 21)
Setting
1. Classifier C with binary output d in {+, -}, a real-valued score s.
1. Instances or data points are generally humans.
2. The + class is desired and the negative - class is not desired.
2. Input X, and
1. one or more sensitive/protected attribute G (e.g. gender) that are part
of the input. E.g. Possible values of G = {m, f}
3. A set of instances sharing a common sensitive attribute is privileged
(receives more + labels).The other is unprivileged (receives less + labels)
4. True output Y
1. Fairness through
Unawareness
• Simple Idea: Do not consider any sensitive attributes when
building the model.
• Advantage: Some support in the law (disparate treatment)?
• Disadvantage:: Other attributes may be correlated with
sensitive attributes (such as job history, geographical location
etc.)
2. Statistical Parity Difference
• Different groups should have the same proportion (or
probability) of positive and negative labels. Ideally the below
value should be close to zero:
• Advantages: Legal support in the form of a rule known as the fourth-fifths rule. May remove
historical bias.
• Disadvantages:
• Trivial classifiers such as classifiers which randomly assign the same of proportion of labels
across different groups satisfy this definition.
• Perfect classifier Y = d may not be allowed if ground truth rates of labels are different across
groups.
3. Equal Opportunity
Difference
• Different groups have the same true positive rate. Ideally the
below value should be close to zero:
• Advantages:
• Perfect classifier allowed.
• Disadvantages:
• May perpetuate historical biases.
• E.g. Hiring application with 100 privileged and 100 unprivileged, but 40 qualified in privileged and 4 in unprivileged.
• By hiring 20 and 2 from each privileged and unprivileged you will satisfy this.
4. False Negative Error
Balance
• If the application is punitive in nature
• Different groups should have the same false negative scores.
• Example:
• The proportion of black defendants who don’t recidivate and receive high risk
scores

Should be the same as
• The proportion of white defendants who don’t recidivate and receive high risk
scores.
5.Test Fairness
• Scores should have the same meaning across different groups.
Impossibility Results
• Core of the debate in COMPAS.
• ProPublica: false negatives should be the same across
different groups
• Northpointe: scores should have the same meaning across
groups. (test fairness)
• Result: If prevalence rates (ground truth proportion of labels
across different groups) are different, and if test fairness is
satisfied then false negatives will differ across groups.
Chouldechova, A., 2017. Fair prediction with disparate impact: A study of bias in recidivism
prediction instruments. Big data, 5(2), pp.153-163.
Tools for Measuring Bias
https://github.com/IBM/AIF360
AI Fairness 360 (AIF 360):
Measuring Bias
Mitigation: Removing Bias
• Mitigation can be happen in three different places:
• Before the model is built, in the training data
• In the model
• After the model is built, with the predictions:
Accuracy = 66%
COMPAS
Before the model is built
• Reweighing (roughly at a high-level):
• Increase weights for some
• Unprivileged with positive labels
• Privileged with negative labels
• Decrease weights for some
• Unprivileged with negative labels
• Privileged with positive labels
+ -
- +
COMPAS
Accuracy = 66%
Accuracy = 66%
Reweighing
AI Fairness 360 Toolkit https://aif360.mybluemix.net
In the model
Zhang, B.H., Lemoine, B. and Mitchell, M., 2018, December. Mitigating
unwanted biases with adversarial learning. In Proceedings of the 2018
AAAI/ACM Conference on AI, Ethics, and Society (pp. 335-340). ACM.
COMPAS
Adversarial
De-biasing
Accuracy = 67%Accuracy = 66%
AI Fairness 360 Toolkit https://aif360.mybluemix.net
After the model is built
• Reject option classification:
• Assume the classifier outputs a probability score.
• If the classifier score is within a small band around 0.5:
• If unprivileged then predict positive
• If privileged then predict negative
Probability of + label for
unprivileged
0 1
0
1
Probability of - label for
unprivileged
COMPAS
Reject
Option
Accuracy = 66% Accuracy = 65%
AI Fairness 360 Toolkit https://aif360.mybluemix.net
Tools
https://github.com/IBM/AIF360
AI Fairness 360 (AIF 360):
Mitigating Bias
Take-home message
• Many forms of fairness and bias exist: most of them are
incompatible with each other.
• Bias can be decreased with algorithms (with usually some
loss in performance)
Thank you
Extras
Choosing Definitions
https://dsapp.uchicago.edu/projects/aequitas/From

Más contenido relacionado

La actualidad más candente

DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictionsAnton Kulesh
 
Explainable AI in Industry (FAT* 2020 Tutorial)
Explainable AI in Industry (FAT* 2020 Tutorial)Explainable AI in Industry (FAT* 2020 Tutorial)
Explainable AI in Industry (FAT* 2020 Tutorial)Krishnaram Kenthapadi
 
Explainable AI
Explainable AIExplainable AI
Explainable AIDinesh V
 
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
 
Responsible AI
Responsible AIResponsible AI
Responsible AINeo4j
 
Fairness in AI (DDSW 2019)
Fairness in AI (DDSW 2019)Fairness in AI (DDSW 2019)
Fairness in AI (DDSW 2019)GoDataDriven
 
Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Hayim Makabee
 
Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...
Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...
Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...Krishnaram Kenthapadi
 
Fairness in Machine Learning and AI
Fairness in Machine Learning and AIFairness in Machine Learning and AI
Fairness in Machine Learning and AISeth Grimes
 
Algorithmic Bias: Challenges and Opportunities for AI in Healthcare
Algorithmic Bias:  Challenges and Opportunities for AI in HealthcareAlgorithmic Bias:  Challenges and Opportunities for AI in Healthcare
Algorithmic Bias: Challenges and Opportunities for AI in HealthcareGregory Nelson
 
Explainability for Natural Language Processing
Explainability for Natural Language ProcessingExplainability for Natural Language Processing
Explainability for Natural Language ProcessingYunyao Li
 
Explainable AI in Healthcare
Explainable AI in HealthcareExplainable AI in Healthcare
Explainable AI in Healthcarevonaurum
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learninginovex GmbH
 
Measures and mismeasures of algorithmic fairness
Measures and mismeasures of algorithmic fairnessMeasures and mismeasures of algorithmic fairness
Measures and mismeasures of algorithmic fairnessManojit Nandi
 
Towards Human-Centered Machine Learning
Towards Human-Centered Machine LearningTowards Human-Centered Machine Learning
Towards Human-Centered Machine LearningSri Ambati
 
Ethical Issues in Machine Learning Algorithms. (Part 3)
Ethical Issues in Machine Learning Algorithms. (Part 3)Ethical Issues in Machine Learning Algorithms. (Part 3)
Ethical Issues in Machine Learning Algorithms. (Part 3)Vladimir Kanchev
 
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)Krishnaram Kenthapadi
 

La actualidad más candente (20)

DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictions
 
Explainable AI in Industry (FAT* 2020 Tutorial)
Explainable AI in Industry (FAT* 2020 Tutorial)Explainable AI in Industry (FAT* 2020 Tutorial)
Explainable AI in Industry (FAT* 2020 Tutorial)
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
 
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
Explainable AIExplainable AI
Explainable AI
 
Responsible AI
Responsible AIResponsible AI
Responsible AI
 
Fairness in AI (DDSW 2019)
Fairness in AI (DDSW 2019)Fairness in AI (DDSW 2019)
Fairness in AI (DDSW 2019)
 
Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)Explainable Machine Learning (Explainable ML)
Explainable Machine Learning (Explainable ML)
 
Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...
Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...
Fairness-aware Machine Learning: Practical Challenges and Lessons Learned (WS...
 
Fairness in Machine Learning and AI
Fairness in Machine Learning and AIFairness in Machine Learning and AI
Fairness in Machine Learning and AI
 
Algorithmic Bias: Challenges and Opportunities for AI in Healthcare
Algorithmic Bias:  Challenges and Opportunities for AI in HealthcareAlgorithmic Bias:  Challenges and Opportunities for AI in Healthcare
Algorithmic Bias: Challenges and Opportunities for AI in Healthcare
 
Explainable AI
Explainable AIExplainable AI
Explainable AI
 
Explainability for Natural Language Processing
Explainability for Natural Language ProcessingExplainability for Natural Language Processing
Explainability for Natural Language Processing
 
Explainable AI in Healthcare
Explainable AI in HealthcareExplainable AI in Healthcare
Explainable AI in Healthcare
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
Measures and mismeasures of algorithmic fairness
Measures and mismeasures of algorithmic fairnessMeasures and mismeasures of algorithmic fairness
Measures and mismeasures of algorithmic fairness
 
Towards Human-Centered Machine Learning
Towards Human-Centered Machine LearningTowards Human-Centered Machine Learning
Towards Human-Centered Machine Learning
 
Ethical Issues in Machine Learning Algorithms. (Part 3)
Ethical Issues in Machine Learning Algorithms. (Part 3)Ethical Issues in Machine Learning Algorithms. (Part 3)
Ethical Issues in Machine Learning Algorithms. (Part 3)
 
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
Responsible AI in Industry (Tutorials at AAAI 2021, FAccT 2021, and WWW 2021)
 

Similar a Explainability and bias in AI

Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...
Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...
Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...Francesca Lazzeri, PhD
 
Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018
Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018
Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018Sri Ambati
 
The Incredible Disappearing Data Scientist
The Incredible Disappearing Data ScientistThe Incredible Disappearing Data Scientist
The Incredible Disappearing Data ScientistRebecca Bilbro
 
The Dangers of Machine Learning
The Dangers of Machine LearningThe Dangers of Machine Learning
The Dangers of Machine LearningtothepointIT
 
​​Explainability in AI and Recommender systems: let’s make it interactive!
​​Explainability in AI and Recommender systems: let’s make it interactive!​​Explainability in AI and Recommender systems: let’s make it interactive!
​​Explainability in AI and Recommender systems: let’s make it interactive!Eindhoven University of Technology / JADS
 
Coder Name Rebecca Oquendo
Coder Name  Rebecca Oquendo                                    Coder Name  Rebecca Oquendo
Coder Name Rebecca Oquendo DioneWang844
 
Human in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AIHuman in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AIPramit Choudhary
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine LearningSri Ambati
 
Coder Name Rebecca Oquendo .docx
Coder Name  Rebecca Oquendo                                    .docxCoder Name  Rebecca Oquendo                                    .docx
Coder Name Rebecca Oquendo .docxmary772
 
The importance of model fairness and interpretability in AI systems
The importance of model fairness and interpretability in AI systemsThe importance of model fairness and interpretability in AI systems
The importance of model fairness and interpretability in AI systemsFrancesca Lazzeri, PhD
 
Explain! Or I will sue you!
Explain! Or I will sue you!Explain! Or I will sue you!
Explain! Or I will sue you!Przemek Biecek
 
Trusted, Transparent and Fair AI using Open Source
Trusted, Transparent and Fair AI using Open SourceTrusted, Transparent and Fair AI using Open Source
Trusted, Transparent and Fair AI using Open SourceAnimesh Singh
 
Adversarial Analytics - 2013 Strata & Hadoop World Talk
Adversarial Analytics - 2013 Strata & Hadoop World TalkAdversarial Analytics - 2013 Strata & Hadoop World Talk
Adversarial Analytics - 2013 Strata & Hadoop World TalkRobert Grossman
 
Using AI to Build Fair and Equitable Workplaces
Using AI to Build Fair and Equitable WorkplacesUsing AI to Build Fair and Equitable Workplaces
Using AI to Build Fair and Equitable WorkplacesData Con LA
 
M2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparencyM2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparencyBoPeng76
 
GDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for Everyone
GDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for EveryoneGDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for Everyone
GDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for EveryoneJames Anderson
 
Towards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptxTowards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptxLuis775803
 
Machine Learning in the Financial Industry
Machine Learning in the Financial IndustryMachine Learning in the Financial Industry
Machine Learning in the Financial IndustrySubrat Panda, PhD
 
Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018HJ van Veen
 
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
 

Similar a Explainability and bias in AI (20)

Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...
Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...
Spark + AI Summit - The Importance of Model Fairness and Interpretability in ...
 
Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018
Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018
Machine Learning Interpretability - Mateusz Dymczyk - H2O AI World London 2018
 
The Incredible Disappearing Data Scientist
The Incredible Disappearing Data ScientistThe Incredible Disappearing Data Scientist
The Incredible Disappearing Data Scientist
 
The Dangers of Machine Learning
The Dangers of Machine LearningThe Dangers of Machine Learning
The Dangers of Machine Learning
 
​​Explainability in AI and Recommender systems: let’s make it interactive!
​​Explainability in AI and Recommender systems: let’s make it interactive!​​Explainability in AI and Recommender systems: let’s make it interactive!
​​Explainability in AI and Recommender systems: let’s make it interactive!
 
Coder Name Rebecca Oquendo
Coder Name  Rebecca Oquendo                                    Coder Name  Rebecca Oquendo
Coder Name Rebecca Oquendo
 
Human in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AIHuman in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AI
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
Coder Name Rebecca Oquendo .docx
Coder Name  Rebecca Oquendo                                    .docxCoder Name  Rebecca Oquendo                                    .docx
Coder Name Rebecca Oquendo .docx
 
The importance of model fairness and interpretability in AI systems
The importance of model fairness and interpretability in AI systemsThe importance of model fairness and interpretability in AI systems
The importance of model fairness and interpretability in AI systems
 
Explain! Or I will sue you!
Explain! Or I will sue you!Explain! Or I will sue you!
Explain! Or I will sue you!
 
Trusted, Transparent and Fair AI using Open Source
Trusted, Transparent and Fair AI using Open SourceTrusted, Transparent and Fair AI using Open Source
Trusted, Transparent and Fair AI using Open Source
 
Adversarial Analytics - 2013 Strata & Hadoop World Talk
Adversarial Analytics - 2013 Strata & Hadoop World TalkAdversarial Analytics - 2013 Strata & Hadoop World Talk
Adversarial Analytics - 2013 Strata & Hadoop World Talk
 
Using AI to Build Fair and Equitable Workplaces
Using AI to Build Fair and Equitable WorkplacesUsing AI to Build Fair and Equitable Workplaces
Using AI to Build Fair and Equitable Workplaces
 
M2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparencyM2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparency
 
GDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for Everyone
GDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for EveryoneGDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for Everyone
GDG Cloud Southlake #17: Meg Dickey-Kurdziolek: Explainable AI is for Everyone
 
Towards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptxTowards Responsible AI - NY.pptx
Towards Responsible AI - NY.pptx
 
Machine Learning in the Financial Industry
Machine Learning in the Financial IndustryMachine Learning in the Financial Industry
Machine Learning in the Financial Industry
 
Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018Hacking Predictive Modeling - RoadSec 2018
Hacking Predictive Modeling - RoadSec 2018
 
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...
 

Más de Bill Liu

Walk Through a Real World ML Production Project
Walk Through a Real World ML Production ProjectWalk Through a Real World ML Production Project
Walk Through a Real World ML Production ProjectBill Liu
 
Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...Bill Liu
 
Productizing Machine Learning at the Edge
Productizing Machine Learning at the EdgeProductizing Machine Learning at the Edge
Productizing Machine Learning at the EdgeBill Liu
 
Transformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to HeroTransformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to HeroBill Liu
 
Deep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps WorkflowsDeep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps WorkflowsBill Liu
 
Metaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at NetflixMetaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at NetflixBill Liu
 
Practical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at ScalePractical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at ScaleBill Liu
 
Building large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudiBuilding large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudiBill Liu
 
Deep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its ApplicationsDeep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its ApplicationsBill Liu
 
Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19Bill Liu
 
Highly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world ApplicationsHighly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world ApplicationsBill Liu
 
Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...Bill Liu
 
Causal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine LearningCausal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine LearningBill Liu
 
Weekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on MobileWeekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on MobileBill Liu
 
Weekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine LearningWeekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine LearningBill Liu
 
AISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with MicroeconomicsAISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with MicroeconomicsBill Liu
 
AISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First WorldAISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First WorldBill Liu
 
AISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the EdgeAISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the EdgeBill Liu
 
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...Bill Liu
 
Toronto meetup 20190917
Toronto meetup 20190917Toronto meetup 20190917
Toronto meetup 20190917Bill Liu
 

Más de Bill Liu (20)

Walk Through a Real World ML Production Project
Walk Through a Real World ML Production ProjectWalk Through a Real World ML Production Project
Walk Through a Real World ML Production Project
 
Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...
 
Productizing Machine Learning at the Edge
Productizing Machine Learning at the EdgeProductizing Machine Learning at the Edge
Productizing Machine Learning at the Edge
 
Transformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to HeroTransformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to Hero
 
Deep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps WorkflowsDeep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps Workflows
 
Metaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at NetflixMetaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at Netflix
 
Practical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at ScalePractical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at Scale
 
Building large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudiBuilding large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudi
 
Deep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its ApplicationsDeep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its Applications
 
Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19
 
Highly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world ApplicationsHighly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world Applications
 
Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...
 
Causal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine LearningCausal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine Learning
 
Weekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on MobileWeekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on Mobile
 
Weekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine LearningWeekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
 
AISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with MicroeconomicsAISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with Microeconomics
 
AISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First WorldAISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First World
 
AISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the EdgeAISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the Edge
 
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
 
Toronto meetup 20190917
Toronto meetup 20190917Toronto meetup 20190917
Toronto meetup 20190917
 

Último

Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxMalak Abu Hammad
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEarley Information Science
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 

Último (20)

Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 

Explainability and bias in AI

  • 1. Explainability and Bias in ML/AI Models Naveen Sundar Govindarajulu August 9, 2019 Visit and sign up RealityEngines.AI
  • 3. COMPAS https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencingFrom: Non recidivating black people twice as likely to be labelled high risk than non recidivating white people
  • 4. Why Explainability? • More use of ML/AI models by laypersons. • Laypersons need explanations • Developers also need quick explanations to debug models faster • There may be a legal need for explanations: • If you deny someone a loan, you may need to explain the reason for the denial.
  • 6. Explainability using Interpretable Models Prior offenses <= 0 Low Risk High Risk Armed offense? Med Risk YES NO NO YES
  • 7. Explainability vs Performance Tradeoff • Some machine learning models are more explainable than others. Performance Explainability Deep learning models Linear Models DecisionTrees
  • 9. What Features? Interpretable Features • We need interpretable features. • Difficult for laypersons to understand raw feature spaces (e.g. word embeddings) • Humans are good at understanding presence or absence of components.
  • 10. Interpretable Instance • E.g. • For Text: • Convert to a binary vector indicating presence or absence of words • For images • Convert to a binary vector indicating presence or absence of pixels or contiguous regions.
  • 11. Method 1: LIME From https://github.com/marcotcr/lime Locally Interpretable Model-agnostic Explanations Ribeiro, M.T., Singh, S. and Guestrin, C., 2016, August. Why Should I Trust You?: Explaining the Predictions of Any Classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135-1144). ACM.
  • 12. Method 1: LIME Any classifier 1 1 0 1 1 0 1 0 0 1 0 0 0 0 1 0 1 1 1 1 0 1 -2.1 1.1 -0.5 2.2 -1.2 -1.5 1 -3 0.8 5.6 1.5 Weights for the linear classifier then give us feature importances Binary vectors -2.1 2.2 -3 5.6 Enforce sparsity
  • 13. Example: Text Sentiment Classification “The movie is not bad” This movie is not bad 0 0 0 2.3 -1.5
  • 14. Explanation for “Cat” LIME with Images From https://github.com/marcotcr/lime
  • 15. Ribeiro, M.T., Singh, S. and Guestrin, C., 2016, August. Why Should I Trust You?: Explaining the Predictions of Any Classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135-1144). ACM. Explanations for Multi-Label Classifiers
  • 16. Ribeiro, M.T., Singh, S. and Guestrin, C., 2016, August. Why Should I Trust You?: Explaining the Predictions of Any Classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135-1144). ACM. Using LIME for Debugging (E.g. 1)
  • 19. Method 2: SHAP Unifies many different feature attribution methods and has some desirable properties. 1. LIME 2. Integrated Gradients 3. Shapley values 4. DeepLift Lundberg, S.M. and Lee, S.I., 2017. A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (pp. 4765-4774).
  • 20. Method 2: SHAP • Derives from game-theoretic foundations. • Shapley values used in game theory to assign values to players in cooperative games.
  • 21. What are Shapley values? • Suppose there is a set S of N players participating in a game with payoff for any S subset of players participating in the game given by: • Shapley values provide one fair way of dividing up the total payoff among the N players.
  • 22. ShapleyValue Payoff for the group including player i Shapley value for player i Payoff for a group without player i
  • 23. SHAP Explanations • Players are features. • Payoff is the model’s real valued prediction.
  • 24. SHAP Implementation (https://github.com/slundberg/shap) Different kinds of explainers: 1. TreeExplainer: fast and exact SHAP values for tree ensembles 2. KernelExplainer: approximate explainer for black box estimators 3. DeepExplainer: high-speed approximate explainer for deep learning models. 4. ExpectedGradients: SHAP-based extension of integrated gradients
  • 25. XGBoost on UCI Income Dataset Output is probability of income over 50k f87 f23 f23 f3 f34 f41 Base ValueOutput
  • 26. Note: SHAP values are Model Dependent. Model 1 Model 2
  • 27. Is This Form of Explainability Enough? • Explainability does not provide us with recourse. • Recourse: Information needed to change a specific prediction to a desired value. • “If you had paid your credit card balance in full for the last three months, you would have got that loan.”
  • 28. Issues with SHAP and LIME SHAP and LIME values are highly variable for instances that are very similar for non-linear models.
 On the Robustness of Interpretability Methods https://arxiv.org/abs/1806.08049
  • 29. Issues with SHAP and LIME SHAP and LIME values are highly variable for instances that are very similar for non-linear models.
 On the Robustness of Interpretability Methods https://arxiv.org/abs/1806.08049
  • 30. Issues with SHAP and LIME SHAP and LIME values don’t provide insight into how the model will behave on new instances. https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16982 High-Precision Model-Agnostic Explanations
  • 31. Take-home message • Explainability is possible need not come at the cost of performance. • Explainability is not enough • Recourse, etc.
  • 32. Bias
  • 33. Fairness and Bias in Machine Learning 1. Bias in this context is unfairness (more or less). 2. Note we are not talking about standard statistical bias in machine learning (the bias in the bias vs. variance tradeoff). 3. For completeness, this is one definition of statistical bias in machine learning. • Bias = Expected value of model - true value
  • 34. Definitions of Fairness or Bias 1. Many, many, many definitions exists. 2. Application dependent. No one definition is better. 3. See “21 Definitions of Fairness” tutorial by Arvind Narayanan,ACM 2018 FAT*. 1. Key Point: Dozens of definitions exist (and not just 21)
  • 35. Setting 1. Classifier C with binary output d in {+, -}, a real-valued score s. 1. Instances or data points are generally humans. 2. The + class is desired and the negative - class is not desired. 2. Input X, and 1. one or more sensitive/protected attribute G (e.g. gender) that are part of the input. E.g. Possible values of G = {m, f} 3. A set of instances sharing a common sensitive attribute is privileged (receives more + labels).The other is unprivileged (receives less + labels) 4. True output Y
  • 36. 1. Fairness through Unawareness • Simple Idea: Do not consider any sensitive attributes when building the model. • Advantage: Some support in the law (disparate treatment)? • Disadvantage:: Other attributes may be correlated with sensitive attributes (such as job history, geographical location etc.)
  • 37. 2. Statistical Parity Difference • Different groups should have the same proportion (or probability) of positive and negative labels. Ideally the below value should be close to zero: • Advantages: Legal support in the form of a rule known as the fourth-fifths rule. May remove historical bias. • Disadvantages: • Trivial classifiers such as classifiers which randomly assign the same of proportion of labels across different groups satisfy this definition. • Perfect classifier Y = d may not be allowed if ground truth rates of labels are different across groups.
  • 38. 3. Equal Opportunity Difference • Different groups have the same true positive rate. Ideally the below value should be close to zero: • Advantages: • Perfect classifier allowed. • Disadvantages: • May perpetuate historical biases. • E.g. Hiring application with 100 privileged and 100 unprivileged, but 40 qualified in privileged and 4 in unprivileged. • By hiring 20 and 2 from each privileged and unprivileged you will satisfy this.
  • 39. 4. False Negative Error Balance • If the application is punitive in nature • Different groups should have the same false negative scores. • Example: • The proportion of black defendants who don’t recidivate and receive high risk scores
 Should be the same as • The proportion of white defendants who don’t recidivate and receive high risk scores.
  • 40. 5.Test Fairness • Scores should have the same meaning across different groups.
  • 41. Impossibility Results • Core of the debate in COMPAS. • ProPublica: false negatives should be the same across different groups • Northpointe: scores should have the same meaning across groups. (test fairness) • Result: If prevalence rates (ground truth proportion of labels across different groups) are different, and if test fairness is satisfied then false negatives will differ across groups. Chouldechova, A., 2017. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big data, 5(2), pp.153-163.
  • 42. Tools for Measuring Bias https://github.com/IBM/AIF360 AI Fairness 360 (AIF 360): Measuring Bias
  • 43. Mitigation: Removing Bias • Mitigation can be happen in three different places: • Before the model is built, in the training data • In the model • After the model is built, with the predictions:
  • 45. Before the model is built • Reweighing (roughly at a high-level): • Increase weights for some • Unprivileged with positive labels • Privileged with negative labels • Decrease weights for some • Unprivileged with negative labels • Privileged with positive labels + - - +
  • 46. COMPAS Accuracy = 66% Accuracy = 66% Reweighing AI Fairness 360 Toolkit https://aif360.mybluemix.net
  • 47. In the model Zhang, B.H., Lemoine, B. and Mitchell, M., 2018, December. Mitigating unwanted biases with adversarial learning. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society (pp. 335-340). ACM.
  • 48. COMPAS Adversarial De-biasing Accuracy = 67%Accuracy = 66% AI Fairness 360 Toolkit https://aif360.mybluemix.net
  • 49. After the model is built • Reject option classification: • Assume the classifier outputs a probability score. • If the classifier score is within a small band around 0.5: • If unprivileged then predict positive • If privileged then predict negative Probability of + label for unprivileged 0 1 0 1 Probability of - label for unprivileged
  • 50. COMPAS Reject Option Accuracy = 66% Accuracy = 65% AI Fairness 360 Toolkit https://aif360.mybluemix.net
  • 52. Take-home message • Many forms of fairness and bias exist: most of them are incompatible with each other. • Bias can be decreased with algorithms (with usually some loss in performance)