STA-395 - Introduction to Machine Learning (Spring 2023)

\(~\)

Syllabus

Welcome to the course 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.

Please note: material will not be posted until we’ve reached that point in the course.

Course Materials

Most class meetings involve both lecture and lab components. Topics are grouped into units, which can be found below. Assignments are generally due at the end of a unit, but I encourage you to complete them as they are made available.

\(~\)

Unit #1

Introduction and Python Foundations

Date	Lecture	Lab
1/24	Introduction to Machine Learning
1/26		Python Foundations

Assignments:

• Lab 1 - due Friday 1/27 by midnight

\(~\)

Unit #2

Machine Learning Concepts and Workflow

Date	Lecture	Lab
1/31	k-Nearest Neighbors	kNN and Pre-processing
2/2	Cross-Validation	Pipelines, Cross-validation, and Tuning Parameters
2/7
2/9	Classifier Performance	Classifier Performance
2/14	Data Preparation	Feature Engineering Challenge

Assignments:

• All labs are due Friday 2/17 by midnight, you may submit one copy per group (make sure both names are included)
• Homework #1 - due Tuesday 2/21 by 1pm

\(~\)

Unit #3

Structured Models and Optimization

Date	Lecture	Lab
2/21	Regression and Gradient Descent Algorithms	Linear Regression and Gradient Descent
2/23	Logistic and Softmax Regression	Logistic and Softmax Regression
2/28	Regularization
3/2	No class (working differently day)
3/7		Lasso, Ridge, and Elastic Net
3/9	Feature Expansion	Discretization, Polynomials, and Splines

Assignments:

• All labs are due Friday 3/10 by midnight, you may submit one copy per group (make sure both names are included)
• Homework #2 - due Tuesday 3/14 by 1pm

\(~\)

Unit #4

Tree-based Models and Ensembles

Date	Lecture	Lab
3/14	Decision Tree Models	Decision Trees
3/16	Random Forests	Ensembles, Stacking, and Random Forests
4/4	Boosting	Boosting and xgboost

Assignments:

• Labs 6 (parts 1 and 2) and 7 (single part) are due Friday 4/7 by midnight
• Optional Lab for 3 points of quiz extra credit if completed by Tuesday 4/4 at 1pm
• Homework #3 - due Tuesday 4/11 by 1pm

Unit #5

Introduction to Deep Learning

Date	Lecture	Lab
4/11	Artificial Neural Networks
4/13		Tensors and PyTorch
4/18	Convolutional Neural Networks	Convolutional Neural Networks
4/20		Transfer Learning
4/25	Recurrent Neural Networks	RNNs in torch

Assignments:

• Lab 8 (parts 1 and 2) due Friday 4/21 by 11:59pm
• Labs 9 and 10 are due Friday 4/28 by 11:59pm
• Homework #4 - due Tuesday 5/2 by 1pm

\(~\)

Unit #6

Introduction to Unsupervised Learning (time permitting, optional lectures and labs)

Date	Lecture	Lab
5/2	Introduction to Unsupervised Learning	Introduction to Unsupervised Learning

• Lab 11 is optional and can be completed for extra credit by Friday 5/5 at 11:59pm

\(~\)

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)
• Intro to Statistical Learning, Ch 2.2 and 2.3 (ML concepts)

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 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’s Intro to Statistical Learning video 6.7 (Lasso)

Unit #4:

• 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 #5:

• Intro to Statistical Learning, Ch 10 (Deep Learning)
• Stanford Engineering Intro to Neural Networks (CS229)
• PyTorch - Learning the Basics
• Stanford’s Intro to Statistical Learning video 10.2 (Convolutional Neural Networks)

Final Project

The last several weeks of the semester will be devoted to a cumulative final project. Intermediate deadlines and reminders will be posted here:

• Upcoming deadline - Plan on presenting a brief 5-minute introduction to your project topic and progress at the start of class on Thursday 5/4.

Guidelines and a detailed description of the project are posted below:

• Link to guidelines