AI For Youth: Grade 09

About Course

This course is an introduction to machine learning. Students learn how a program learns from examples instead of rules, how data becomes features and labels, how to run a first model by hand, and how to tell whether it worked. Features: self-paced, no calculus or coding, about 45 minutes per lesson, 8 lessons, 8 quizzes, 1 project, no prerequisites.

Course Content

Unit 1: Learning from Data
<p>The shift from writing rules to collecting examples, how a dataset becomes a table of features and labels, a first model you can run by hand, and how to tell whether it actually worked. Lessons 1 through 4.</p>

Lesson 1: Rules versus Examples
Quiz 1: Rules versus Examples
Lesson 2: Features and Labels
Quiz 2: Features and Labels
Lesson 3: Nearest Neighbor
Quiz 3: Nearest Neighbor
Lesson 4: Accuracy and the Train/Test Split
Quiz 4: Accuracy and the Train/Test Split

Unit 2: Models and Their Limits
<p>Memorising instead of learning, the three kinds of task a model can be set, where training data comes from and what it carries with it, and a capstone. Lessons 5 through 8.</p>

Lesson 5: Overfitting and Underfitting
Quiz 5: Overfitting and Underfitting
Lesson 6: Classification, Regression and Clustering
Quiz 6: Classification, Regression and Clustering
Lesson 7: Data Sources and Bias
Quiz 7: Data Sources and Bias
Lesson 8: Capstone — Building a Predictor by Hand
Quiz 8: Building a Predictor by Hand
Capstone Project: Build a Predictor by Hand

About Course

Course Content

Unit 1: Learning from Data <p>The shift from writing rules to collecting examples, how a dataset becomes a table of features and labels, a first model you can run by hand, and how to tell whether it actually worked. Lessons 1 through 4.</p>

Lesson 1: Rules versus Examples

Quiz 1: Rules versus Examples

Lesson 2: Features and Labels

Quiz 2: Features and Labels

Lesson 3: Nearest Neighbor

Quiz 3: Nearest Neighbor

Lesson 4: Accuracy and the Train/Test Split

Quiz 4: Accuracy and the Train/Test Split

Unit 2: Models and Their Limits <p>Memorising instead of learning, the three kinds of task a model can be set, where training data comes from and what it carries with it, and a capstone. Lessons 5 through 8.</p>

Lesson 5: Overfitting and Underfitting

Quiz 5: Overfitting and Underfitting

Lesson 6: Classification, Regression and Clustering

Quiz 6: Classification, Regression and Clustering

Lesson 7: Data Sources and Bias

Quiz 7: Data Sources and Bias

Lesson 8: Capstone — Building a Predictor by Hand

Quiz 8: Building a Predictor by Hand

Capstone Project: Build a Predictor by Hand

Welcome to Sheaves AI

Unit 1: Learning from Data
<p>The shift from writing rules to collecting examples, how a dataset becomes a table of features and labels, a first model you can run by hand, and how to tell whether it actually worked. Lessons 1 through 4.</p>

Unit 2: Models and Their Limits
<p>Memorising instead of learning, the three kinds of task a model can be set, where training data comes from and what it carries with it, and a capstone. Lessons 5 through 8.</p>