John Hopkins University - Exploring Algorithmic Bias as a Policy Issue: A Teach-Out
- Offered byCoursera
Exploring Algorithmic Bias as a Policy Issue: A Teach-Out at Coursera Overview
Duration | 9 hours |
Start from | Start Now |
Total fee | Free |
Mode of learning | Online |
Official Website | Explore Free Course  |
Credential | Certificate |
Exploring Algorithmic Bias as a Policy Issue: A Teach-Out at Coursera Highlights
- Earn a certificate after completion of the course
- Assessment and quizzes
- Financial aid available
Exploring Algorithmic Bias as a Policy Issue: A Teach-Out at Coursera Course details
- This Teach Out does not issue certificates of completion.
- Algorithms and algorithmic bias are making regular appearances in the news, and increasingly, are being recognized as a policy issue. But what is an algorithm, exactly? And what does it mean when someone describes an algorithm as biased?
- This Teach-Out will encourage policy makers, agency leaders, and others in similar positions to identify algorithms that are already in use and make connections to broader ideas about fairness, justice, and equity. After completing the Teach-Out, learners will be able to participate in discussions around algorithmic bias, inform others about how algorithms can perpetuate existing disparities, and take steps to reduce the impact of algorithmic bias on the people and communities they serve.
Exploring Algorithmic Bias as a Policy Issue: A Teach-Out at Coursera Curriculum
Welcome to the Course
What is an Algorithm?
Defining Algorithm
Digital and Analog Algorithms
Examples of Algorithms
Complexity and Digitization
Related Concepts: Artificial Intelligence, Machine Learning, and Automation
Key Takeaways
Common Types of Algorithms
Common Ways Algorithms Operate
The Algorithm Development Process
Examples of Where Algorithms Are Being Used
Key Takeaways
A note about language
A guide to spotting algorithms "in the wild"
How to Recognize AI Snake Oil (Arvind Narayanan)
Optional: Eighteen pitfalls to beware of in AI Journalism (Sayash Kapoor and Arvind Narayanan)
Why It So Hard to Regulate Algorithms (Todd Feathers, The Markup)
Optional reading: Machine learning, explained (Sara Brown, MIT Sloan School of Management)
Optional reading: Why AI is just automation (Joshua A. Kroll, Brookings Institute)
How Does the Public Sector Identify Problems It Tries to Solve with AI? (Maria Levy Daniel, Tech Policy Press)
Resource: Common Terms and their Definitions
AI can be a force for good or ill in society, so everyone must shape it, not just the tech guys (Afua Bruce, The Guardian)
Activity: Building an Algorithms List
Test yourself: Is it an algorithm
Reflection
What Does It Mean for an Algorithm To Be Biased?
When We Say Bias What Does That Mean?
Related Terms: Fairness, Equality, Equity, and Justice
Different Ways of Assessing Algorithms
Examples of Algorithmic Bias
Key Takeaways
Overview of Sources of Algorithmic Bias
Sources of Bias in Problem Definition
Sources of Bias in Data
Sources of Bias in Algorithm Development
Sources of Bias in Use
Key Takeaways
Optional: Origins of Algorithmic Bias
Machine Bias (Julia Angwin, Jeff Larson, Surya Mattu, & Lauren Kirchner, ProPublica)
Activity: Explore the Survival of the Best Fit Game
Technology is Biased Too. How Do We Fix It? (Laura Hudson, FiveThirtyEight)
Reclaiming the Stories that Algorithms Tell (David G. Robinson, O’Reilly)
Exploring Different Contexts for AI Policy
Activity: Identifying sources of bias
Reflections
Algorithmic Bias and Systemic Bias
Who Decides What Kinds of Problems Algorithms Should Solve?
Who Decides What Gets Measured and How?
Who Decides When and How an Algorithm Should be Created and Used?
Who Decides When and How Algorithms Should Be Regulated?
Who Gets to Opt Out?
Who Gets to Be an Expert?
Lydia X.Z. Brown: Algorithms cannot be separated from the context of creation
Sarah A. Riley: Unintended consequences beyond systematically disadvantaging marginalized groups
Sarah A. Riley: Impact on People of Color
Lydia X.Z. Brown: Impact on Disabled People
Lydia X.Z. Brown: Algorithmic bias is already here
Sarah A. Riley: Implementation is a choice. Be accountable.
Key Takeaways: Algorithms, Bias, and Power
Systemic Racism in AI: How Algorithms Replicate White Supremacy and Injustice (Bunny McKensie Mack , Teen Vogue)
AI Creators Want Us to Believe AI Is an Existential Threat. Why? (Ryan Calo, Undark)
These Women Tried to Warn Us About AI (Neil, Rolling Stone)
Activity: Connecting current algorithms to historic choices
Reflections
Anticipating and Addressing Algorithmic Bias
Increasing Participatory Methods in All Phases of Algorithmic Development
Addressing Bias in Problem Definition
Addressing Bias in Measurement and Data
Addressing Bias in Model Creation and Algorithm Design
Addressing Bias in How Algorithms Are Used
Key Takeaways
Ensure That Everyone Is Working From the Same Understanding of Algorithms
Reference Specific Details in How Algorithms Are Developed and Used
Introduce Others to the Idea of Algorithmic Bias
Share Information About How Algorithmic Bias Occurs
Connect Algorithmic Bias to Systemic Power
Work Toward Specific Solutions That Address Algorithmic Bias
Key Takeaways
Optional Reading: Resources on algorithmic accountability and assessment
A menu of strategies to address algorithmic bias
Activity: Building an algorithmic fairness toolkit
We Let Tech Companies Frame the Debate Over AI Ethics. That Was a Mistake. (Robert Hart, Undark)
Getting Beyond Minimizing Harms of Algorithmic Systems (Tech Policy Press)
The White House AI R&D Strategy Offers a Good Start Here How to Make it Better (Sarah Myers West, Tech Policy Press)
Optional readings: Current and proposed legislation
Relevant Organizations Working on Algorithmic Bias and Related Issues
Activity: Identify groups/organizations working on algorithmic bias issues
Closing Thoughts
Algorithmic Bias Conversation Starters
Areas of Discussion Where Policy Experts Can Contribute
Reflection
Exploring Algorithmic Bias as a Policy Issue: A Teach-Out at Coursera Admission Process
Important Dates
Other courses offered by Coursera
Student Forum
Anything you would want to ask experts?
Data ScienceDeep LearningArtificial IntelligenceExploring Algorithmic Bias as a Policy Issue: A Teach-Out