CS 445: Machine Learning
James Madison University, Fall 2021

Course Information

Location: EnGeo 2209
Meeting Time: T/TH 11:20-12:35PM
Prerequisites: A grade of "C-" or better in CS 327 and MATH 220 or equivalent
Course Web Page: http://w3.cs.jmu.edu/spragunr/CS445/
Required Textbook: Introduction to Data Mining (Second Edition), Pang-Ning Tan et. al. Pearson, 2019.

Instructor Information

Name: Dr. Nathan Sprague
Office: ISAT/CS 226
Office Phone: 568-3312
Email: spragunr@jmu.edu
Office Hours: http://w3.cs.jmu.edu/spragunr/schedule.html

Questions related to course content should be asked through Piazza. You may use email if you need to contact me directly.

You are welcome to call or stop by my office any time, with the understanding that I may or may not be available outside of my posted office hours.

I'll be experimenting with a hybrid online/in-person format for office hours this semester. You are free to attend in-person or online via Zoom. The Zoom link will be made available on the course Canvas page. I reserve the right to modify the office hour format if the hybrid approach turns out to be unworkable. If you need to meet with me outside of my posted hours, email me to make an appointment.

Course Description

In this course, you will be introduced to machine learning, a class of techniques which involve algorithms that "learn by example". Machine learning is prevalent in many fields: autonomous driving, detecting credit card fraud and cyber attacks, and organizing/searching through the ever-growing set of photos on your phone.

Machine learning algorithms can be used to solve problems that would be difficult or impossible to solve using hand-written code. For example, imagine writing a traditional program to decode a zip code from a picture of an envelope. Machine learning uses labeled examples of hand-written zip codes (digits) and "learns" from these examples to recognize/decode zip codes from millions of pieces of mail each day. This course focuses on a balance of theoretical and practical knowledge. Popular methods such as neural networks, deep learning, and support vector machines will be covered. Small projects will allow students to apply these techniques and communicate results in a quantitative manner.

Course Goals

At the conclusion of this course students should:

Course Content and Preliminary Schedule

A detailed schedule, including reading and homework assignments, is available on the course web page. That schedule will be updated throughout the semester and should be checked regularly.

Methods of Evaluation

Course grades will be based on two midterms and a final exam, in-class activities, homework assignments, projects, and occasional quizzes. Projects will include 4-5 programming assignments and a poster presentation. Assignment specifications and due dates will be posted to the course schedule page. The final grade will be computed as follows:

In-Class Activities 15%
Reading Quizzes 5%
Muddiest Points 5%
Projects and Homework Assignments 20%
Final Poster Project 10%
Midterms 25%
Final Exam 20%

Letter grades will be assigned on the scale A=90-100, B=80-89, C=70-79, D=60-69, F=0-59, with potential minor adjustments after considering the overall performance of the class and actual distribution of numeric scores. I will use "+" and "-" grades at my discretion. I do not assign WP or WF grades except under extraordinary circumstances.

Muddiest Points

Most weeks I will ask you to submit a "muddiest point" reflection highlighting concepts that were unclear from the reading or lecture.

Course Policies

Attendance and Participation

Regular attendance and fully engaged participation is expected. Your grade will be partially based on in-class assignments and quizzes, so attendance will affect your grade.

Academic Integrity

It is expected that your work in this course will comply with the provisions of the JMU honor code: http://www.jmu.edu/honor/code.shtml. It is not a violation of the honor code to discuss assignments with other students. However, all individual work that you submit must be written by you, based on your own understanding of the material. Representing someone else's work as your own, in any form, constitutes an honor code violation. It is also a violation of the honor code to "render unauthorized assistance to another student by knowingly permitting him or her to see or copy all or a portion of an examination or any work to be submitted for academic credit."

A key component of academic integrity is giving credit where credit is due. If you receive assistance, either from another student or from some other source, you must explicitly acknowledge that fact in your submission.

I will prosecute honor code violations if they come to my attention. If in doubt about what is allowed, ask me.

Missed and Late Assignment Policy

If you are unable to take an exam at the scheduled time because of illness or other problems, you must contact me beforehand to arrange to take the exam at a different time. Failure to make prior arrangements for a missed exam will result in a grade of 0 for the exam.

It will not be possible to receive credit for in-class work that is missed due to absence. In recognition of the fact that absences are occasionally unavoidable, I will drop the two lowest scores in this category when calculating your final grade.

Homework and Programming assignments will be due at 11:00PM on the posted due-date. Assignments submitted after the deadline will be subject to a 25% penalty for the first day. The penalty will increase by 25% every 24 hours. No assignments will be accepted more than 48 hours after the deadline.

Unless there are extraordinary circumstances, I will not provide extensions for illnesses or other personal difficulties. Instead, you will have three "late days" which may be applied to any project or homework assignment (with some exceptions for assignments due at the end of the semester or immediately before exams). The use of a late day allows you to extend the deadline for an assignment by 24 hours. You may apply multiple late days to a single assignment, but they may not be used to submit assignments more than 48 hours late.

Classroom Behavior

Phones should be silenced and put away during class unless they are being used for a course-related activity. If you violate this policy, I reserve the right to give you a long pointless lecture about the failings of your generation and the probable downfall of civilization. This will be embarrassing for both of us.

Please be ready to begin work at the scheduled start of class and refrain from packing up to leave until the class period is over. In exchange, I will commit to ending class promptly at the schedule time.

Adding/Dropping

Students are responsible for adding and dropping courses via MyMadison. Please consult the registrar's page of dates and deadlines for exact deadlines.

Disability Accommodations

If you need an accommodation based on the impact of a disability, you should contact the Office of Disability Services (Student Success Center, Room 1202, www.jmu.edu/ods, 540-568-6705) if you have not previously done so. Disability Services will provide you with an Access Plan Letter that will verify your need for services and make recommendations for accommodations to be used in the classroom. Once you have presented me with this letter, you and I will sit down and review the course requirements, your disability characteristics, and your requested accommodations to develop an individualized plan, appropriate for this course.

Inclement Weather Policy

This class will operate in accord with JMU's inclement weather policy available at http://www.jmu.edu/JMUpolicy/1309.shtml

Religious Observation Accommodations

I will give reasonable and appropriate accommodations to students requesting them on grounds of religious observation. If you require such accommodations you must notify me at least two weeks in advance.

Acknowledgments

Some of the course content this semester (including portions of this Syllabus) are borrowed from Kevin Molloy's Fall 2019 and Fall 2020 Machine Learning Courses.