CSCI 4930/5750 Machine Learning
This section of the course (Spring 2018) meets Mondays, Wednesdays, and Fridays from 2:10 to 3:00PM in Ritter Hall, Room 202.
This course introduces students to the field of machine learning with emphasis on the probabilistic models that dominate contemporary applications. Students will discover how computers can learn from examples and extract salient patterns hidden in large data sets. The course will introduce classification algorithms that predict discrete states for variables as well as regression algorithms that predict continuous values for variables. Attention will be given to both supervised and unsupervised settings in which (respectively) labeled training data is or is not available. Emphasis is placed on both the conceptual relationships between these different learning problems as well as the statistical models and computational methods used to employ those models.
- What is Machine Learning?
- Classification algorithms (decision trees, naive Bayes)
- Clustering algorithms (hierarchical, k-Means)
- Recommendation Engines (kNN)
- Non-linear methods (kernel methods, SVMs)
- Neural Networks, Deep Learning
- Noisy Channel Model in NLP
- Word alignment, EM algorithm
- n-gram language modeling
- HMMs, Viterbi
- Learning from Graphs
- Ethics in Machine Learning
Student Learning OutcomesAfter successfully completing this course, students will be able to:
- select a machine learning algorithm and model appropriate for a given problem, and apply an existing implementation to a real dataset;
- formulate an appropriate evaluation scheme in order to tune model parameters and evaluate a solution;
- implement at least two machine learning algorithms from scratch
- (CSCI 5750) apply machine learning techniques to a solve a research problem in the student's major field
For those of you who choose to use the lab computers, please read the department and university policies on appropriate use of computer systems.
Homework and Exams
I will give ~6 in-class quizzes (roughly one every two weeks) throughout the semester; dates TBA, depending on our progress through the course material. The quizzes are usually true/false, multiple choice, and some short answer, and only take about 10-15 minutes at the beginning of class. I'll drop your lowest quiz score, but I will not allow you to make up quizzes that you miss because of absence or if you arrive late for class. Together the quizzes make up 20% of your final grade.
There will be a 50-minute in-class midterm exam on Wednesday, March 7th, worth 15% of your final grade, and a final exam Wednesday, May 9th, from 2:00-3:50PM, worth 25% of your final grade.
You will also be asked to do a semester software project related to some topic we cover in the course, accounting for 20% of your final grade. I'll give you some ideas as we approach the middle of the semester. Since we cover a lot of different things, this is a good opportunity for you to explore some particular topic in greater depth.
Finally, we'll have a machine learning "bakeoff" toward the end of the semester, in which you will solve a fixed machine learning problem of my choosing. I'll provide you with labeled training data for you to learn from, and then we will evaluate your algorithm against a hidden test set. Prizes will be awarded for the best submissions. The bakeoff will count for 20% of your final grade.
- Student percentage above 93% will result in a grade of A or better.
- Student percentage above 90% will result in a grade of A- or better.
- Student percentage above 87% will result in a grade of B+ or better.
- Student percentage above 84% will result in a grade of B or better.
- Student percentage above 80% will result in a grade of B- or better.
- Student percentage above 77% will result in a grade of C+ or better.
- Student percentage above 74% will result in a grade of C or better.
- Student percentage above 70% will result in a grade of C- or better.
- Student percentage above 60% will result in a grade of D or better.
- Student percentage below 60% will result in a grade of F.
Academic Integrity Statement
Academic integrity is honest, truthful and responsible conduct in all academic endeavors. The mission of Saint Louis University is "the pursuit of truth for the greater glory of God and for the service of humanity." Accordingly, all acts of falsehood demean and compromise the corporate endeavors of teaching, research, health care, and community service via which SLU embodies its mission. The University strives to prepare students for lives of personal and professional integrity, and therefore regards all breaches of academic integrity as matters of serious concern. The governing University-level Academic Integrity Policy was adopted in Spring 2015, and can be accessed on the Provost's Office website. Additionally, each SLU College, School, and Center has adopted its own academic integrity policies, available on their respective websites. All SLU students are expected to know and abide by these policies, which detail definitions of violations, processes for reporting violations, sanctions, and appeals. Please direct questions about any facet of academic integrity to your faculty, the chair of the department of your academic program, or the Dean/Director of the College, School or Center in which your program is housed.
Title IX Statement
Saint Louis University and its faculty are committed to supporting our students and seeking an environment that is free of bias, discrimination, and harassment. If you have encountered any form of sexual misconduct (e.g. sexual assault, sexual harassment, stalking, domestic or dating violence), we encourage you to report this to the University. If you speak with a faculty member about an incident of misconduct, that faculty member must notify SLU’s Title IX coordinator and share the basic fact of your experience. The Title IX coordinator will then be available to assist you in understanding all of your options and in connecting you with all possible resources on and off campus. If you wish to speak with a confidential source, you may contact the counselors at the University Counseling Center at 314-977-TALK. To view SLU’s sexual misconduct policy and for resources, please visit this web address.
Student Success Center
In recognition that people learn in a variety of ways and that learning is influenced by multiple factors (e.g., prior experience, study skills, learning disability), resources to support student success are available on campus. The Student Success Center, a one-stop shop, which assists students with academic and career related services, is located in the Busch Student Center (Suite, 331) and the School of Nursing (Suite, 114). Students who think they might benefit from these resources can find out more about (1) Course-level support (e.g., faculty member, departmental resources, etc.) by asking your course instructor, and (2) University-level support (e.g., tutoring services, university writing services, disability services, academic coaching, career services, and/or facets of curriculum planning) by visiting the Student Success Center or by going here.
Disability Services Academic Accommodations
Students with a documented disability who wish to request academic accommodations are encouraged to contact Disability Services to discuss accommodation requests and eligibility requirements. Please contact Disability Services, located within the Student Success Center, at <Disability_services@slu.edu> or 314-977-3484 to schedule an appointment. Confidentiality will be observed in all inquiries. Once approved, information about academic accommodations will be shared with course instructors via email from Disability Services and viewed within Banner via the instructor’s course roster.