STA-395 - Introduction to Machine Learning (Spring ’24)

\(~\)

Welcome to the website for Sta-395, Introduction to Machine Learning! To begin, you can find the course syllabus linked below:

• Click here to view the course syllabus

You can locate course content by scrolling, or by using the navigation bar in the upper-left.

Course Materials

Most class meetings involve both lecture and lab components. Topics are organized into units, which can be found below. Assignment submissions should be made via P-web following the instructions that are provided. Labs are generally due on Fridays at midnight, while homework is due at the start of the class session when we begin the next unit.

\(~\)

Unit #1

Overview, Introductory Python, and Unsupervised Learning:

Date	Lecture	Lab
1/23	Introduction and k-means	Lab 1 - Python Foundations and k-means
1/25		Continue Lab 1
1/30	Hierarchical clustering and DBSCAN	Lab 2 - More Clustering
2/1	Principal Components (guest lecture)	Lab 3 - Principal Components

Assignments:

• Introduction to Python and \(k\)-means clustering lab (Lab 1) is due Friday 1/26 at 11:59pm
  • Please submit a Jupyter Notebook via P-web containing all code and written responses
• The clustering and outlier detection lab (Lab 2), as well as the principal components lab (Lab 3) are due Friday 2/2 at 11:59pm
  • Please submit a Jupyter Notebook via P-web containing all code and written responses
• Homework 1 due Thursday 2/8 at 1:00pm
  • Please submit a Jupyter Notebook via P-web containing all code and written responses

\(~\)

Unit #2

Machine Learning Workflow and Classical Models:

Date	Lecture	Lab
2/6	Training, Validation, and KNN	Lab 4 - Workflow and pipelines
2/8		Continue Lab 4
2/13	Cross-validation	Lab 5 - Cross-validation
2/15	Classifier Performance	Lab 6 - Classifier Performance
2/20	Decision Trees	Lab 7 - Decision Trees
2/22	Support Vector Machines	Lab 8 - SVMs
2/27		Finish labs 7 and 8

Assignments:

• Lab 4 is due on Tuesday 2/13 at 11:59pm
• Labs 5 and 6 are due on Tuesday 2/20 Thursday 2/22 at 11:59pm
• Labs 7 and 8 are due on Friday 2/23 Tuesday 2/27 at 11:59pm
• Homework 2 due Tuesday 2/27 Thursday 2/29 at 1pm
  • Please submit a Jupyter Notebook via P-web containing all code and written responses

\(~\)

Unit #3

Ensembles:

Date	Lecture	Lab
2/29	Random Forests	Lab 9 - Ensembles and Stacking
3/5	Boosting	Lab 10 - XGBoost
3/7		Lab 11 - Feature Engineering

• Labs 9 and 10 are due Friday 3/8 at 11:59pm
• Lab 11 is due Friday 3/15 at 11:59pm
• Homework 3 is due Tuesday 4/2 at 11:59pm
• Project groups are due Monday 3/11 at 11:59pm (share via email)

\(~\)

Unit #4

Structured Models and the Mathematics of “Learning”:

Date	Lecture	Lab
3/14		Linear Algebra and Calculus Review (optional)
4/2	Linear Regression and Gradient Descent Algorithms	Lab 12 - Linear Regression and Gradient Descent
4/4	Logistic and Softmax Regression	Lab 13 - Logistic Regression and Stochastic Gradient Descent
4/9	Regularization and Regression	Lab 14 - Lasso, Ridge, and Elastic Net
4/11	Artificial Neural Networks
4/16	Review/catch up
4/18	Exam	Exam

• The optional lab (linear and algebra and calculus review) is due Tuesday 4/2
• Labs 12 and 13 are due Tuesday 4/9 at 11:59pm
• Lab 14 is due Friday 4/12 at 11:59pm
• Homework 4 is due Wednesday 4/17 at 11:59pm
• Study materials for the midterm exam can be found below:
  • Study guide
  • Practice exam
    • Sketch solution

\(~\)

Unit #5

Introduction to Deep Learning:

Date	Lecture	Lab
4/23		Lab 15 - Introduction to PyTorch
4/25	Convolutional Neural Networks	Lab 16 - CNNs in PyTorch
4/30		Lab 17 - Transfer Learning using PyTorch
5/2	Reccurent Neural Networks	Lab 18 - RNNs in PyTorch (optional)
5/7	Presentations (part 1)
5/9	Presentations (part 2)

• There will be no homework assignment during this unit so that you can prioritize working on the project outside of class
  • Due to scheduling constraints, we will not be having project briefings on 4/25 and you’ll be asked to share a short update via email instead
    • Please include a Jupyter notebook and a brief written explanation of your progress in your update
• All labs from Unit 5 are due Friday 5/10 at 11:59pm

Final Project

Below is the assignment page describing the final project:

• Project assignment description

\(~\)

Supplemental Resources

Unit #1:

• UW-Madison SSCC Python Data Wrangling tutorials (balanced)
• Python Data Science Handbook Chapters 2,3,4 (more technical)
• Stefanie Molin’s Pandas workshop tutorials 1, 2 (less technical)

Unit #2:

• Intro to Statistical Learning, Ch 2.2 and 3.5 (kNN)
• Intro to ML with Python Ch 1 and Ch 2.1 (kNN)
• Stanford’s Intro to Statistical Learning video 5.2 (k-fold CV)
• Stanford’s Intro to Statistical Learning video 5.3 (CV the right and wrong way)
• Intro to ML with Python Ch 5 (Model Evaluation and Improvement)

Unit #3:

• Intro to Statistical Learning, Ch 8 (Decision Tree Models)
• Stanford’s Intro to Statistical Learning video 8.4 (Bagging and Random Forest)
• Stanford’s Intro to Statistical Learning video 8.5 (Boosting)
• Google’s Intro to Gradient Boosted Trees

Unit #4:

• Intro to ML with Python Ch 2 (Linear Models)
• Intro to Statistical Learning, Ch 4.2 and 4.3 (Logistic Regression)
• Intro to Statistical Learning, Ch 6.2 (Regularization)
• Stanford Engineering Intro to Neural Networks (CS229)

Unit #5:

• PyTorch - Learning the Basics
• Stanford’s Intro to Statistical Learning video 10.2 (Convolutional Neural Networks)