Applied Machine Learning: Ensemble Learning

Explore how to make powerful, accurate predictions with ensemble learners, one of the most common classes of machine learning algorithms.

Introduction

• Explore ensemble learning
• What you should know
• What tools you need
• Using the exercise files

1. Review Machine Learning Basics

• What is machine learning?
• What does machine learning look like in real life?
• What does an end-to-end machine learning pipeline look like?
• Bias-Variance trade-off

2. Preparing the Data

• Reading in the data
• Cleaning up continuous features
• Cleaning up categorical features
• Write out all train, validation, and test sets

3. What is Ensemble Learning?

• What is ensemble learning?
• How does ensemble learning work?
• Why is ensemble learning so powerful?

4. Boosting

• What is boosting?
• How does boosting reduce overall error?
• When should you consider using boosting?
• What are examples of algorithms that use boosting?
• Explore boosting algorithms in Python
• Implement a boosting model

5. Bagging

• What is bagging?
• How does bagging reduce overall error?
• When should you consider using bagging?
• What are examples of algorithms that use bagging?
• Explore bagging algorithms in Python
• Implement a bagging model

6. Stacking

• What is stacking?
• How does stacking reduce overall error?
• When should you consider using stacking?
• What are examples of algorithms that use stacking?
• Explore stacking algorithms in Python
• Implement a stacking model

Conclusion

• Compare the three methods
• Compare all models on validation set
• How to continue advancing your skills