Evaluating Machine Learning Models
Students learn various metrics and techniques for evaluating the performance and robustness of machine learning models.
Key Questions
- How do we measure the success or failure of an intelligent system using appropriate metrics?
- Differentiate between overfitting and underfitting in machine learning models.
- Justify the selection of specific evaluation metrics based on the problem context.
Common Core State Standards
About This Topic
This topic explores the controversial policy of Affirmative Action, the use of race-conscious policies to address historical discrimination and promote diversity. Students examine the legal history from Regents v. Bakke to the most recent Supreme Court rulings. They analyze the tension between the goal of 'substantive equality' (remedying past wrongs) and 'procedural equality' (colorblindness).
For seniors, this is a highly relevant topic as they apply to college. It requires them to engage with complex ethical questions about fairness, merit, and the role of institutions in shaping a diverse society. This topic comes alive when students can physically model the patterns of admissions and hiring through a simulated committee process where they must balance competing values.
Active Learning Ideas
Simulation Game: The Admissions Committee
Students act as a college admissions board. They are given 'profiles' of applicants with varying grades, backgrounds, and life experiences. They must decide who to admit while following (or challenging) specific diversity goals.
Formal Debate: Colorblind vs. Race-Conscious
Students debate the 'Justice Roberts' view ('The way to stop discrimination on the basis of race is to stop discriminating on the basis of race') versus the 'Justice Sotomayor' view (that race still matters in American life).
Gallery Walk: Affirmative Action Around the World
Display how other countries (like India or South Africa) handle 'reservations' or 'quotas' for marginalized groups. Students compare these systems to the US 'holistic review' model.
Watch Out for These Misconceptions
Common MisconceptionColleges use 'quotas' to meet diversity goals.
What to Teach Instead
Quotas were ruled unconstitutional in the 1978 Bakke case. Peer investigations into 'Holistic Review' help students understand that race can be a 'plus factor' but not a fixed number or a separate track.
Common MisconceptionAffirmative action only benefits Black students.
What to Teach Instead
Historically, white women have been the largest beneficiaries of affirmative action policies in employment. Peer-led 'Demographic Data Analysis' helps students see the broad impact of these policies across different groups.
Suggested Methodologies
Ready to teach this topic?
Generate a complete, classroom-ready active learning mission in seconds.
Frequently Asked Questions
What is 'Strict Scrutiny' in the context of race?
What did the Supreme Court rule in the 2023 Harvard/UNC cases?
How can active learning help students understand Affirmative Action?
What are 'Race-Neutral' alternatives?
More in Data Science and Intelligent Systems
Introduction to Data Science Workflow
Students learn the end-to-end process of data science, from data acquisition and cleaning to analysis and communication of results.
2 methodologies
Big Data Concepts and Pattern Recognition
Students analyze massive datasets to find hidden trends, using statistical libraries to process and visualize complex information sets.
2 methodologies
Data Visualization and Interpretation
Students learn to create effective data visualizations to communicate insights and identify patterns in complex datasets.
2 methodologies
Fundamentals of Machine Learning: Supervised Learning
Students are introduced to supervised learning, exploring concepts like regression and classification and how models learn from labeled data.
2 methodologies
Fundamentals of Machine Learning: Unsupervised Learning
Students explore unsupervised learning techniques like clustering and dimensionality reduction to find hidden structures in unlabeled data.
2 methodologies