CSIS 441 · Intro to ML

CSIS 441 Machine Learning & Computational Modeling

Course Description

An introduction to machine learning with application. The course covers the fundamentals of classification and regression, as well as, the methodology behind the design and implementing various models. Students will be introduced to a variety of topics including parameter estimation, model training/fitting, goodness of fit, generalization, regularization, inference, and objective & loss functions. Along the way, students will learn practical tools for machine learning & statistical building models, applications of machine learning & statistical models, and related mathematics.

Instructor

J. Walker Orr, Ph.D.
Office hours: WMR 216 (see schedule)

Texts

required

Peter Bruce, Andrew Bruce, & Peter Gedeck Practical Statistics for Data Scientists ISBN-13: 978-1492072942

Resources

Objectives

The goal of this course is to provide you with the foundational knowledge of machine learning & computational modeling and its applications. To achieve this goal, you will study the underlying mathematics, statistics, and algorithms to used to create machine learning models. You will learn to appropriately apply these methods to solve complex, real-world problems.

Course Organization

The course will include regular homework and/or programming assignments. Unless otherwise specified, assignments are due 5 minutes before midnight on the due date. There will be no credit given for late assignments (without an excused absence)—turn in as much as you can.

Reading assignments should be completed before the lecture covering the material. Not all reading material will be covered in the lectures, but you will be responsible for the material on homework and exams. Quizzes over the assigned reading may be given at any time.

Collaboration

See the GFU CS/IS/Cyber policies for collaboration and discussion of collaboration and academic integrity. Most students would be surprised at how easy it is to detect collaboration in programming—please do not test us! Remember: you always have willing and legal collaborators in the faculty.

Almost all of life is filled with collaboration (i.e., people working together). Yet in our academic system, we artificially limit collaboration. These limits are designed to force you to learn fundamental principles and build specific skills. It is very artificial but intensional for your own benefit. The only way for you to learn is by doing the work.

To be clear, do not:

Copy other students work
Copy code from an online sources such as Stack Overflow
Generate code with "AI" tools such as ChatGPT or GitHub Copilot

Spiritual Formation

Besides EYS, I am always available to discuss the Christian faith if you have any questions or doubts. Send me an email, come by my office hours, or talk to my after class, Christ is the reason I am at GFU, I always have time to talk about faith.

Grading

The final course grade will be based on:

Grading Scale

60% Homework & programming assignments
20% Final Project
10% Participation
10% Midterm

Tentative Schedule

Week 1

Introduction & Statistics

Reading: Chapter 1: pages 1 – 46

1/20

MLK, Jr. Holiday—no classes

Reading: –

Week 2

Sampling & Distributions

Reading: Chapter 2: 47 – 86

Week 3

Classification & Regression

Reading: First half of chapter 3: pages 87 – 113; & chapter 4: pages 141 – 154

Week 4

Linear Model Details

Reading: Chapter 4: pages 155 – 173

2/14

Mid-semester break—no classes

Week 5

Model Fitting

Reading: Chapter 5: pages 195 – 200 & pages 221 – 230

Week 6

SGD

Reading: Chapter 5: pages 195 – 200 & pages 221 – 230

Week 7

Decision Boundaries

Reading: Chapter 5: pages 208 – 215

Week 8

Hypothesis Spaces

3/12

Midterm exam

Week 9

K Nearest Neighbors & Non-linearity

Reading: Chapter 6: pages 237 – 248

Week 10

Decision Trees

Reading: Chapter 6: pages 249 – 258

Week 11

Spring Break

Week 12

Neural Networks & Regularization

Reading: Chapter 6 & Supplementary Materials

Week 13

More Neural Networks: Deep Learning