Debugging Bias: The Lab
Learning Objective
By the end of this lesson, you will be able to audit a specific AI model to identify the source of algorithmic bias and propose mitigation strategies to fix it using the FATP framework.
📚 Key Concepts
- • Auditing: Examining AI systems for bias
- • Mitigation: Fixing identified problems
- • Edge Case: Unusual scenarios that reveal flaws
🎯 What You'll Do
- • Discover bias in an AI application
- • Investigate training data problems
- • Apply the FATP framework
- • Create mitigation strategies