Introduction to Machine Learning

Instructor:
Bruce MacLennan [he/his/him]
Phone: 974-0994
Office: Min Kao 550
Hours: WF 2:30–3:30, or make an appointment
Email: maclennan@utk.edu

GTAs:
Nicholas Van Nostrand
Office: Min Kao 217
Hours: MWF 1:00–2:00, or make an appointment
Email: nvannost at vols.utk.edu

Michael Price
Office: Programming Clinic (Min Kao 416)
Hours: TR 11:00–12:30, or make an appointment
Email: mprice35 at vols.utk.edu 

Classes: 1:25–2:15 MWF in Min Kao 404

This page: http://web.eecs.utk.edu/~mclennan/Classes/425-528

Information

• Description
• Prerequisites
• Grading
• Text
• Student Learning Outcomes [pdf]
• Accommodations
• Tentative List of Topics
• Additional Information
  
  
• MacLennan’s home page
• Department of Electrical Engineering & Computer Science
  
• UTK Home Page

Description

COSC 425:

Machine learning is concerned with computer programs that automatically improve their performance through experience. This course covers the theory and practice of machine learning from a variety of perspectives. We cover topics such as learning decision trees, neural network learning, statistical learning methods, genetic algorithms, Bayesian learning methods, explanation-based learning, and reinforcement learning. Programming assignments include hands-on experiments with various learning algorithms.

COSC 528:

Theoretical and practical aspects of machine learning techniques that enable computer systems to learn from experience. Methods studied include concept learning, decision tree learning, neural networks, Bayesian learning, instance-based learning, genetic algorithms, rule learning, analytical learning, and reinforcement learning.

N.B.

This is an updated version of COSC 425/528. A Tentative List of Topics is below, but it may be adjusted during the semester.

Prerequisites

COSC 425:

(RE) Prerequisite(s): 302; Electrical and Computer Engineering 313 or Mathematics 323. Comment(s): Prior knowledge may satisfy prerequisite with consent of instructor.

COSC 528: 

Recommended Background: 302, 311, Mathematics 251; and Mathematics 323, or Electrical and Computer Engineering 313.

N.B.

You will be programming at least six machine learning algorithms. I recommend python or Matlab/Octave, but C++ and Java are also acceptable. If you want to use another language, please contact me.

Grading

• Homework (occasional)
• Projects (five)
• The projects will count for 90% and the homeworks for the remaining 10%
  
• Late policy: As competent software engineers, you are expected to hand in assignments on time.
  One day late: –20%, two days last: –40%, three or more days late: no credit. No exceptions!
• Tentative Assignment Schedule 
• No exams
  
• Subject to change!
  

Text

Accommodations

For Students with Disabilities

Students who have a disability that requires accommodation(s) should make an appointment with the Office of Disability Services (974-6087) to discuss their specific needs as well as schedule an appointment with me during my office hours.

Name and Pronoun Accommodations

If you use a name and/or pronouns other than what is in the course roll, please email me with the name and/or pronouns that you would like me to use and I will be glad to accommodate this request.

Tentative List of Topics

We will spend a week or two on each of these topics, which correspond to the indicated chapters in the textbook.

1. Introduction (ch. 1)
   
2. Supervised Learning (ch. 2)
   
3. Bayesian Decision Theory (ch. 3)
4. Parametric Methods (chs. 4–5)
5. Dimensionality Reduction (ch. 6)
6. Clustering (ch. 7)
7. Non-Parametric Methods (ch. 8)
8. Decision Trees (ch. 9)
9. Neural Networks (chs. 10–11)
10. Local Models (ch. 12)
11. Kernel Machines (ch. 13)
12. Reinforcement Learning (ch. 18)
13. Machine Learning Experiments (ch. 19)
    

Additional Information

•  Resources for projects 

Send mail to Bruce MacLennan / MacLennan@utk.edu

This page in web.eecs.utk.edu/~mclennan/Classes/425-528
Last updated:  2018-11-09.