SlideShare a Scribd company logo
Fairness in Automated Decision Systems
Data Protectionin Software Development- WS 21/22
Dr. AmirshayanAhmadian
• Sai Sri Praveen Gadiyaram
• Sowjanya Chennamaneni
26-01-2022 1
OUTLINE
1. Motivation
2. Background
3. Challenges
4. Fairness metrics
5. Fairness approaches
6. Fairness Architecture
7. Adversarial Debiasing
8. Case-study
9. Summary
26-01-2022 2
MOTIVATION
26-01-2022 3
BACKGROUND
26-01-2022
AutomatedDecision Systems : Decisionstaken by the systems without any human involvement.
4
.
26-01-2022
Name Education Race Gender Zip code
John Bachelor's White Male 56783
Mary Bachelor's White Female 45673
Annie Bachelor's Black Female 15674
Jack Bachelor's Black Male 13456
TrainedML
model
Predictions
Eligible
Yes
Yes
No
No
BACKGROUND
5
• Protected Attributes: Can be a source of social bias. For example, age, race and gender
• Unprotected Attributes: Attributes that are not protected.
• Fairness ThroughAwareness:
An algorithm is fair if it gives similar predictions to similar individuals.
It is also known as Individual Fairness.
• DemographicParity :
The likelihood of a positive outcome should be the same regardless of whether the person is in
the protected (e.g., female) group. It comes under Group Fairness
26-01-2022 6
BACKGROUND
CHALLENGES
• Decisions are dependent on the "Data".
• A model trained on biased data can produce
unfair and discriminatory decisions.
• Algorithm bias
26-01-2022
Interpretabilityof models
Universalfairnessdefinition
Universalfairnessmetrics
Concretebias mitigation
Lack of social fairness
Data Algorithm Decision
7
FAIRNESS METRICS
26-01-2022
S. No Metric Definition formula
1. Disparate Impact
Ratio(DIR)
Ratio of the rate of positive outcome for
minority group to majority group
P(y=1|x=0)/P(y=1|x=1)
2. Statistical Parity
Difference(SPD)
Difference of the rate of positive outcome for
minority group to majority group
P(y=1|x=0) - P(y=1|x=1)
• Quantifying Group Fairness
• 2 Groups:
• Male, x=1
• Female, x=0
8
FAIRNESS METRICS
• Effectively search the attribute space
• Test-intensive – Can't be done manually
• Generate automatic Test cases for
indiscrimination.
26-01-2022
Protected
Non-
protected
Individual
discrimination
• path coverage and individual discrimination
• Always negate a non-protected attribute
• Success score= |setof generated testcasesfrom unprotected attributes|
|individual discrimination detected|
9
FAIRNESS APPROACHES
• Achieving Fairness == Mitigating Bias
• Consider fairness constraints in decision system building
• 3 different approaches to include fairness constraints
26-01-2022
Pre-processing
• Learn fair
Representations
• Remove
disparate impact
In-processing
• Regularization
• Train
independently
and tune
together
• Adversarial
debiasing
Post-processing
• Calibrated Equal
Odds
10
FAIRNESS ARCHITECTURE
26-01-2022 11
ADVERSARIAL DE-BIASING
• Optimize with respect to the fairness loss
• Adversarial NN can predict the attributes which leads to an unfair decision.
26-01-2022
Gender ??
Race ??
12
CASE-STUDY
• Fairness architecture at LinkedIn
• "Equal opportunities to equally
qualified members"
• Model agnostic and scalable
• Fair model analyzer:
• Specialized test set
• Creates the fairness correction
• Mitigation trainer
• Learns from the previous step output
• post-processing approach
26-01-2022 13
SUMMARY
26-01-2022
• 100 % data-driven systems
are not ideal
• Focus on Bias
• Data source & variety
• Regularization
• Lack of Fairness knowledge
• Social Fairness, ethics
• Fairness constraints
• Context & diversity
• Continuous fairness evaluation
• Testing after deployment
• Optimize fairness metrics
14
REFERENCES
• Mehrabi, Ninareh & Morstatter, Fred & Saxena, Nripsuta & Lerman, Kristina & Galstyan, Aram. (2019). A Survey on Bias and Fairness in Machine Learning.
• Kulshrestha, Ankit and Ilya Safro. “CONFAIR: Configurable and Interpretable Algorithmic Fairness.” (2021).
• https://engineering.linkedin.com/blog/2022/a-closer-look-at-how-linkedin-integrates-fairness-into-its-ai-pr
• https://www.slideshare.net/godatadriven/fairness-in-ai-ddsw-2019
• https://blog.insightdatascience.com/tackling-discrimination-in-machine-learning-5c95fde95e95
• https://www.persistent.com/blogs/fairness-in-ai-systems/
• Aniya Aggarwal, Pranay Lohia, Seema Nagar, Kuntal Dey, and Diptikalyan Saha. 2019. Black box fairness testing of machine learning models. In <i>Proceedings of the
2019 27th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering</i> (<i>ESEC/FSE
2019</i>). Association for Computing Machinery, New York, NY, USA, 625–635. DOI:https://doi.org/10.1145/3338906.3338937
• Ignore icons created by Parzival’ 1997 - Flaticon
• Ethics icons created by Freepik - Flaticon
• Testing icons created by surang – Flaticon
• https://www.datanami.com/2018/09/05/how-to-build-a-better-machine-learning-pipeline/
• https://www.oreilly.com/library/view/ai-fairness/9781492077664/ch01.html
• https://fortune.com/2018/10/10/amazon-ai-recruitment-bias-women-sexist
26-01-2022 15
26-01-2022
Thank you.
16

More Related Content

Similar to Fairness in Automated Decision Systems

Caveon Webinar Series - Lessons Learned at the 2015 National Conference on S...
Caveon Webinar Series -  Lessons Learned at the 2015 National Conference on S...Caveon Webinar Series -  Lessons Learned at the 2015 National Conference on S...
Caveon Webinar Series - Lessons Learned at the 2015 National Conference on S...
Caveon Test Security
 
PSY 239 401 Chapter 5 SLIDES
PSY 239 401 Chapter 5 SLIDESPSY 239 401 Chapter 5 SLIDES
PSY 239 401 Chapter 5 SLIDES
kimappel
 
Ch2 POM.pdf chapter summary for principal of management
Ch2 POM.pdf chapter summary for principal of managementCh2 POM.pdf chapter summary for principal of management
Ch2 POM.pdf chapter summary for principal of management
abbasgass48
 
What is the impact of bias in data analysis, and how can it be mitigated.pdf
What is the impact of bias in data analysis, and how can it be mitigated.pdfWhat is the impact of bias in data analysis, and how can it be mitigated.pdf
What is the impact of bias in data analysis, and how can it be mitigated.pdf
Soumodeep Nanee Kundu
 
Decisions
DecisionsDecisions
Decisions
shafaqsheikh
 
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
 
Ch._6_pp_industrial.ppt
Ch._6_pp_industrial.pptCh._6_pp_industrial.ppt
Ch._6_pp_industrial.ppt
MichelleFlynn36
 
Successful adoption of Machine Learning
Successful adoption of Machine LearningSuccessful adoption of Machine Learning
Successful adoption of Machine Learning
Rudradeb Mitra
 
Learning Analytics in Higher Education
Learning Analytics in Higher EducationLearning Analytics in Higher Education
Learning Analytics in Higher Education
Jose Antonio Omedes
 
Trends in machine learning (1)
Trends in machine learning (1)Trends in machine learning (1)
Trends in machine learning (1)
Viswanath Shenoi
 
Algorithmic Fairness: A Brief Introduction
Algorithmic Fairness: A Brief IntroductionAlgorithmic Fairness: A Brief Introduction
Algorithmic Fairness: A Brief Introduction
AnthonyMelson
 
Decision Making process.ppt
Decision Making process.pptDecision Making process.ppt
Decision Making process.ppt
FaheemGhazanfar1
 
Fair AI
Fair AIFair AI
MIS & DSS
MIS & DSSMIS & DSS
Statistics — Your Friend, Not Your Foe
Statistics — Your Friend, Not Your Foe Statistics — Your Friend, Not Your Foe
Statistics — Your Friend, Not Your Foe
Integrity Management Services, Inc.
 
Decision making systems
Decision making systemsDecision making systems
Decision making systems
Shwetabh Jaiswal
 
Responsible Data Use in AI - core tech pillars
Responsible Data Use in AI - core tech pillarsResponsible Data Use in AI - core tech pillars
Responsible Data Use in AI - core tech pillars
Sofus Macskássy
 
Management Concepts and Meaning and more
Management Concepts and Meaning and moreManagement Concepts and Meaning and more
Management Concepts and Meaning and more
JIMS Rohini Sector 5
 
Management Concepts and Meaning-Decision Making
Management Concepts and Meaning-Decision MakingManagement Concepts and Meaning-Decision Making
Management Concepts and Meaning-Decision Making
JIMS Rohini Sector 5
 
Online Data Science Competitions(Kaggle)- Pranav Bahl
Online Data Science Competitions(Kaggle)- Pranav BahlOnline Data Science Competitions(Kaggle)- Pranav Bahl
Online Data Science Competitions(Kaggle)- Pranav Bahl
Pranav Bahl
 

Similar to Fairness in Automated Decision Systems (20)

Caveon Webinar Series - Lessons Learned at the 2015 National Conference on S...
Caveon Webinar Series -  Lessons Learned at the 2015 National Conference on S...Caveon Webinar Series -  Lessons Learned at the 2015 National Conference on S...
Caveon Webinar Series - Lessons Learned at the 2015 National Conference on S...
 
PSY 239 401 Chapter 5 SLIDES
PSY 239 401 Chapter 5 SLIDESPSY 239 401 Chapter 5 SLIDES
PSY 239 401 Chapter 5 SLIDES
 
Ch2 POM.pdf chapter summary for principal of management
Ch2 POM.pdf chapter summary for principal of managementCh2 POM.pdf chapter summary for principal of management
Ch2 POM.pdf chapter summary for principal of management
 
What is the impact of bias in data analysis, and how can it be mitigated.pdf
What is the impact of bias in data analysis, and how can it be mitigated.pdfWhat is the impact of bias in data analysis, and how can it be mitigated.pdf
What is the impact of bias in data analysis, and how can it be mitigated.pdf
 
Decisions
DecisionsDecisions
Decisions
 
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)
 
Ch._6_pp_industrial.ppt
Ch._6_pp_industrial.pptCh._6_pp_industrial.ppt
Ch._6_pp_industrial.ppt
 
Successful adoption of Machine Learning
Successful adoption of Machine LearningSuccessful adoption of Machine Learning
Successful adoption of Machine Learning
 
Learning Analytics in Higher Education
Learning Analytics in Higher EducationLearning Analytics in Higher Education
Learning Analytics in Higher Education
 
Trends in machine learning (1)
Trends in machine learning (1)Trends in machine learning (1)
Trends in machine learning (1)
 
Algorithmic Fairness: A Brief Introduction
Algorithmic Fairness: A Brief IntroductionAlgorithmic Fairness: A Brief Introduction
Algorithmic Fairness: A Brief Introduction
 
Decision Making process.ppt
Decision Making process.pptDecision Making process.ppt
Decision Making process.ppt
 
Fair AI
Fair AIFair AI
Fair AI
 
MIS & DSS
MIS & DSSMIS & DSS
MIS & DSS
 
Statistics — Your Friend, Not Your Foe
Statistics — Your Friend, Not Your Foe Statistics — Your Friend, Not Your Foe
Statistics — Your Friend, Not Your Foe
 
Decision making systems
Decision making systemsDecision making systems
Decision making systems
 
Responsible Data Use in AI - core tech pillars
Responsible Data Use in AI - core tech pillarsResponsible Data Use in AI - core tech pillars
Responsible Data Use in AI - core tech pillars
 
Management Concepts and Meaning and more
Management Concepts and Meaning and moreManagement Concepts and Meaning and more
Management Concepts and Meaning and more
 
Management Concepts and Meaning-Decision Making
Management Concepts and Meaning-Decision MakingManagement Concepts and Meaning-Decision Making
Management Concepts and Meaning-Decision Making
 
Online Data Science Competitions(Kaggle)- Pranav Bahl
Online Data Science Competitions(Kaggle)- Pranav BahlOnline Data Science Competitions(Kaggle)- Pranav Bahl
Online Data Science Competitions(Kaggle)- Pranav Bahl
 

Recently uploaded

Webinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for EmbeddedWebinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for Embedded
ICS
 
KuberTENes Birthday Bash Guadalajara - Introducción a Argo CD
KuberTENes Birthday Bash Guadalajara - Introducción a Argo CDKuberTENes Birthday Bash Guadalajara - Introducción a Argo CD
KuberTENes Birthday Bash Guadalajara - Introducción a Argo CD
rodomar2
 
Fundamentals of Programming and Language Processors
Fundamentals of Programming and Language ProcessorsFundamentals of Programming and Language Processors
Fundamentals of Programming and Language Processors
Rakesh Kumar R
 
Unveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdfUnveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdf
brainerhub1
 
How to write a program in any programming language
How to write a program in any programming languageHow to write a program in any programming language
How to write a program in any programming language
Rakesh Kumar R
 
E-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian Companies
E-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian CompaniesE-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian Companies
E-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian Companies
Quickdice ERP
 
Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)
Julian Hyde
 
Transform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR SolutionsTransform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR Solutions
TheSMSPoint
 
Automated software refactoring with OpenRewrite and Generative AI.pptx.pdf
Automated software refactoring with OpenRewrite and Generative AI.pptx.pdfAutomated software refactoring with OpenRewrite and Generative AI.pptx.pdf
Automated software refactoring with OpenRewrite and Generative AI.pptx.pdf
timtebeek1
 
What is Master Data Management by PiLog Group
What is Master Data Management by PiLog GroupWhat is Master Data Management by PiLog Group
What is Master Data Management by PiLog Group
aymanquadri279
 
socradar-q1-2024-aviation-industry-report.pdf
socradar-q1-2024-aviation-industry-report.pdfsocradar-q1-2024-aviation-industry-report.pdf
socradar-q1-2024-aviation-industry-report.pdf
SOCRadar
 
AI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI App
AI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI AppAI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI App
AI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI App
Google
 
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit ParisNeo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j
 
Using Query Store in Azure PostgreSQL to Understand Query Performance
Using Query Store in Azure PostgreSQL to Understand Query PerformanceUsing Query Store in Azure PostgreSQL to Understand Query Performance
Using Query Store in Azure PostgreSQL to Understand Query Performance
Grant Fritchey
 
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
mz5nrf0n
 
GreenCode-A-VSCode-Plugin--Dario-Jurisic
GreenCode-A-VSCode-Plugin--Dario-JurisicGreenCode-A-VSCode-Plugin--Dario-Jurisic
GreenCode-A-VSCode-Plugin--Dario-Jurisic
Green Software Development
 
A Study of Variable-Role-based Feature Enrichment in Neural Models of Code
A Study of Variable-Role-based Feature Enrichment in Neural Models of CodeA Study of Variable-Role-based Feature Enrichment in Neural Models of Code
A Study of Variable-Role-based Feature Enrichment in Neural Models of Code
Aftab Hussain
 
Using Xen Hypervisor for Functional Safety
Using Xen Hypervisor for Functional SafetyUsing Xen Hypervisor for Functional Safety
Using Xen Hypervisor for Functional Safety
Ayan Halder
 
SMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API ServiceSMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API Service
Yara Milbes
 
E-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet DynamicsE-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet Dynamics
Hornet Dynamics
 

Recently uploaded (20)

Webinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for EmbeddedWebinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for Embedded
 
KuberTENes Birthday Bash Guadalajara - Introducción a Argo CD
KuberTENes Birthday Bash Guadalajara - Introducción a Argo CDKuberTENes Birthday Bash Guadalajara - Introducción a Argo CD
KuberTENes Birthday Bash Guadalajara - Introducción a Argo CD
 
Fundamentals of Programming and Language Processors
Fundamentals of Programming and Language ProcessorsFundamentals of Programming and Language Processors
Fundamentals of Programming and Language Processors
 
Unveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdfUnveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdf
 
How to write a program in any programming language
How to write a program in any programming languageHow to write a program in any programming language
How to write a program in any programming language
 
E-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian Companies
E-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian CompaniesE-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian Companies
E-Invoicing Implementation: A Step-by-Step Guide for Saudi Arabian Companies
 
Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)
 
Transform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR SolutionsTransform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR Solutions
 
Automated software refactoring with OpenRewrite and Generative AI.pptx.pdf
Automated software refactoring with OpenRewrite and Generative AI.pptx.pdfAutomated software refactoring with OpenRewrite and Generative AI.pptx.pdf
Automated software refactoring with OpenRewrite and Generative AI.pptx.pdf
 
What is Master Data Management by PiLog Group
What is Master Data Management by PiLog GroupWhat is Master Data Management by PiLog Group
What is Master Data Management by PiLog Group
 
socradar-q1-2024-aviation-industry-report.pdf
socradar-q1-2024-aviation-industry-report.pdfsocradar-q1-2024-aviation-industry-report.pdf
socradar-q1-2024-aviation-industry-report.pdf
 
AI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI App
AI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI AppAI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI App
AI Fusion Buddy Review: Brand New, Groundbreaking Gemini-Powered AI App
 
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit ParisNeo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
 
Using Query Store in Azure PostgreSQL to Understand Query Performance
Using Query Store in Azure PostgreSQL to Understand Query PerformanceUsing Query Store in Azure PostgreSQL to Understand Query Performance
Using Query Store in Azure PostgreSQL to Understand Query Performance
 
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
 
GreenCode-A-VSCode-Plugin--Dario-Jurisic
GreenCode-A-VSCode-Plugin--Dario-JurisicGreenCode-A-VSCode-Plugin--Dario-Jurisic
GreenCode-A-VSCode-Plugin--Dario-Jurisic
 
A Study of Variable-Role-based Feature Enrichment in Neural Models of Code
A Study of Variable-Role-based Feature Enrichment in Neural Models of CodeA Study of Variable-Role-based Feature Enrichment in Neural Models of Code
A Study of Variable-Role-based Feature Enrichment in Neural Models of Code
 
Using Xen Hypervisor for Functional Safety
Using Xen Hypervisor for Functional SafetyUsing Xen Hypervisor for Functional Safety
Using Xen Hypervisor for Functional Safety
 
SMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API ServiceSMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API Service
 
E-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet DynamicsE-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet Dynamics
 

Fairness in Automated Decision Systems

  • 1. Fairness in Automated Decision Systems Data Protectionin Software Development- WS 21/22 Dr. AmirshayanAhmadian • Sai Sri Praveen Gadiyaram • Sowjanya Chennamaneni 26-01-2022 1
  • 2. OUTLINE 1. Motivation 2. Background 3. Challenges 4. Fairness metrics 5. Fairness approaches 6. Fairness Architecture 7. Adversarial Debiasing 8. Case-study 9. Summary 26-01-2022 2
  • 4. BACKGROUND 26-01-2022 AutomatedDecision Systems : Decisionstaken by the systems without any human involvement. 4
  • 5. . 26-01-2022 Name Education Race Gender Zip code John Bachelor's White Male 56783 Mary Bachelor's White Female 45673 Annie Bachelor's Black Female 15674 Jack Bachelor's Black Male 13456 TrainedML model Predictions Eligible Yes Yes No No BACKGROUND 5 • Protected Attributes: Can be a source of social bias. For example, age, race and gender • Unprotected Attributes: Attributes that are not protected.
  • 6. • Fairness ThroughAwareness: An algorithm is fair if it gives similar predictions to similar individuals. It is also known as Individual Fairness. • DemographicParity : The likelihood of a positive outcome should be the same regardless of whether the person is in the protected (e.g., female) group. It comes under Group Fairness 26-01-2022 6 BACKGROUND
  • 7. CHALLENGES • Decisions are dependent on the "Data". • A model trained on biased data can produce unfair and discriminatory decisions. • Algorithm bias 26-01-2022 Interpretabilityof models Universalfairnessdefinition Universalfairnessmetrics Concretebias mitigation Lack of social fairness Data Algorithm Decision 7
  • 8. FAIRNESS METRICS 26-01-2022 S. No Metric Definition formula 1. Disparate Impact Ratio(DIR) Ratio of the rate of positive outcome for minority group to majority group P(y=1|x=0)/P(y=1|x=1) 2. Statistical Parity Difference(SPD) Difference of the rate of positive outcome for minority group to majority group P(y=1|x=0) - P(y=1|x=1) • Quantifying Group Fairness • 2 Groups: • Male, x=1 • Female, x=0 8
  • 9. FAIRNESS METRICS • Effectively search the attribute space • Test-intensive – Can't be done manually • Generate automatic Test cases for indiscrimination. 26-01-2022 Protected Non- protected Individual discrimination • path coverage and individual discrimination • Always negate a non-protected attribute • Success score= |setof generated testcasesfrom unprotected attributes| |individual discrimination detected| 9
  • 10. FAIRNESS APPROACHES • Achieving Fairness == Mitigating Bias • Consider fairness constraints in decision system building • 3 different approaches to include fairness constraints 26-01-2022 Pre-processing • Learn fair Representations • Remove disparate impact In-processing • Regularization • Train independently and tune together • Adversarial debiasing Post-processing • Calibrated Equal Odds 10
  • 12. ADVERSARIAL DE-BIASING • Optimize with respect to the fairness loss • Adversarial NN can predict the attributes which leads to an unfair decision. 26-01-2022 Gender ?? Race ?? 12
  • 13. CASE-STUDY • Fairness architecture at LinkedIn • "Equal opportunities to equally qualified members" • Model agnostic and scalable • Fair model analyzer: • Specialized test set • Creates the fairness correction • Mitigation trainer • Learns from the previous step output • post-processing approach 26-01-2022 13
  • 14. SUMMARY 26-01-2022 • 100 % data-driven systems are not ideal • Focus on Bias • Data source & variety • Regularization • Lack of Fairness knowledge • Social Fairness, ethics • Fairness constraints • Context & diversity • Continuous fairness evaluation • Testing after deployment • Optimize fairness metrics 14
  • 15. REFERENCES • Mehrabi, Ninareh & Morstatter, Fred & Saxena, Nripsuta & Lerman, Kristina & Galstyan, Aram. (2019). A Survey on Bias and Fairness in Machine Learning. • Kulshrestha, Ankit and Ilya Safro. “CONFAIR: Configurable and Interpretable Algorithmic Fairness.” (2021). • https://engineering.linkedin.com/blog/2022/a-closer-look-at-how-linkedin-integrates-fairness-into-its-ai-pr • https://www.slideshare.net/godatadriven/fairness-in-ai-ddsw-2019 • https://blog.insightdatascience.com/tackling-discrimination-in-machine-learning-5c95fde95e95 • https://www.persistent.com/blogs/fairness-in-ai-systems/ • Aniya Aggarwal, Pranay Lohia, Seema Nagar, Kuntal Dey, and Diptikalyan Saha. 2019. Black box fairness testing of machine learning models. In <i>Proceedings of the 2019 27th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering</i> (<i>ESEC/FSE 2019</i>). Association for Computing Machinery, New York, NY, USA, 625–635. DOI:https://doi.org/10.1145/3338906.3338937 • Ignore icons created by Parzival’ 1997 - Flaticon • Ethics icons created by Freepik - Flaticon • Testing icons created by surang – Flaticon • https://www.datanami.com/2018/09/05/how-to-build-a-better-machine-learning-pipeline/ • https://www.oreilly.com/library/view/ai-fairness/9781492077664/ch01.html • https://fortune.com/2018/10/10/amazon-ai-recruitment-bias-women-sexist 26-01-2022 15