**Course title: ****Machine Learning**

**Course number: **COMP.5450 (graduate), COMP.4220 (undergraduate)

**Semester: **Spring 2017

**Location: **Olsen Hall 503

**Meeting Times: **Tue., 5:30PM-8:15PM

**Instructor: **Dr. Jerome J. Braun, jbraun@cs.uml.edu

**TA: **Xinzi Sun, Xinzi_Sun@student.uml.edu

**Course o****verview**

This introductory course gives an overview of machine learning techniques used in data mining and pattern recognition applications. Topics include: foundations of machine learning, including statistical and probabilistic methods; generative and discriminative models; linear regression; Bayesian methods; parametric and non-parametric classification; supervised and unsupervised learning; clustering and dimensionality reduction; anomaly detection; and applications to very large datasets.

**Prerequisites**

This is a graduate/upper-level undergraduate course. All students should have be familiar with fundamentals of probability, linear algebra, and calculus. All must also be familiar with algorithmic and programming techniques needed for the implementation of machine-learning algorithms that will be taught in this course.

The above proficiencies are assumed typical for students who have successfully completed the following courses (or their equivalents): Discrete Structures II (or Linear Algebra), Probability & Statistics I, Computing II.

Students enrolled in the graduate section are expected to have proficiency in the areas of probability and linear algebra commensurate with their graduate status, and are assumed typical for students who have completed Linear Algebra I and II and Calculus III.

**Quiz 0 on prerequisite math knowledge **(Probability & Statistics, Calculus, basic Matrix Algebra) will be given in the first week of class. This quiz tests mathematical skills necessary for the course. Moreover, this quiz is intended to help the students self-evaluate their prior knowledge of mathematical fundamentals expected of students taking this course. In all cases, the responsibility to have or to acquire mathematical skills needed for this course shall rest with the students.

**Textbooks**

Required textbook:

- Bishop, C. M., ‘
*Pattern Recognition and Machine Learning*,’ Springer, 2006.

Other recommended books:

- Koller, D. and Friedman, N. ‘
*Probabilistic Graphical Models*,’ MIT Press. 2009. - Hastie, T., Tibshirani, R. and Friedman, J., ‘
*The Elements of Statistical Learning*,’ Springer. 2001. - Theodoridis, S. and Koutroumbas K., ‘
*Pattern Recognition*,’ Academic Press. - Duda, R.O., Hart, P.E., and Stork, D.G., ‘
*Pattern Classification*,’ Wiley-Interscience, 2001. - Russell, S. and Norvig, N., ‘
*Artificial Intelligence: A Modern Approach*,’ Prentice Hall Series in Artificial Intelligence. 2003. - Scholkopf B. and Smola A., ‘
*Learning with Kernels*,’ MIT Press, 2002. - Vapnik, V., ‘
*The Nature of Statistical Learning Theory*,’ Springer. - Goodfellow, I., Bengio, Y., Courville, A., ‘
*Deep Learning*,’ MIT Press, 2016.

**Topics Overview (tentative)**

- Introduction to Machine Learning. Review of mathematical concepts.
- Probabilistic Approaches. Maximum Likelihood Principle.
- Multivariate linear regression. Bayesian regression. Bias-variance trade-off.
- Logistic regression, One-vs-all classification, Regularization.
- Artificial Neural Networks.
- Practical advice for applying learning algorithms: debugging, feature/model design, performance evaluation methods.
- Support Vector Machines (SVMs).
- Hidden Markov Models (HMM).
- Boosting and Bootstrapping.
- Unsupervised learning: clustering.
- Introduction to deep learning and large-scale machine learning.

**Student Learning Outcomes**

After successfully completing this course, the students should be able to understand:

- Design, construction and evaluation of a machine-learning (pattern-recognition) system.
- Mathematical foundations of several mainstream machine-learning algorithms.
- Major classes of approaches in machine-learning (pattern-recognition) literature.
- Selected theoretical issues involved in machine-learning algorithm design.
- Implementation of machine-learning techniques in scientific-computing environment.

These goals will be evaluated through quizzes, homework assigments, and a final project.

**Deliverables/Graded work**

Students will be evaluated based on the following graded work:

- In-class quizzes: Quiz 0 (5%), Quiz 1 (20%), Quiz 2 (20%).
- Should a midterm (40%) or a final (40%) quiz/exam be announced
*in lieu*(*instead*) of Quizzes 1 and 2, such an announcement would be made in advance. - Homework problem-sets, up to 4 assignments: 10% total.
- Self-graded; must hand in completed homework and grade; self-assigned grades will be verified by us at random.
- Late-homework Policy: 50% off any assignment handed in up to one week after its due date (i.e., by the beginning of the following Tuesday lecture); 100% off (no credit) afterwards.
- Final Project (40%).
- Class attendance and participation (5%).

Graduate-section students are expected to achieve higher scores and/or submit additional work to receive the same grade as undergraduate-section students.

**Policies:**

**Academic Honesty Policy**: Students are expected to honor and follow all CS department and UMass Lowell policies related to academic honesty and integrity. Violators risk failing the course in addition to any actions taken by university administration. Cheating will not be tolerated, and students who cheat risk failing the course and possible university administrative actions.

All work on exams must be the student's own work.

The work on homework assignments must also be the student's own work, with the following exceptions: 1) hints provided by the TA or the instructor may be used, provided that after obtaining such hints the students perform the assignment on their own, and that having obtained hints is acknowledged in writing in the student's work; 2) forming study-groups is allowed (and encouraged) and students may engage in discussions related to homework assignments, provided that following such discussions students complete the homework assignment separately on their own (without referring/copying detailed notes from those discussions) and that occurrence of such discussions is acknowledged in writing on the homework assignment (however, doing a homework assignment together by more than one person is not permitted). Using homework solutions from any source, such as websites or past-year’s solutions obtained from any source, is not permitted.

**Violations may result in a warning, a possible grade of 0, and/or a report to the university.**

**Project work: **The mandatory required final-project (see graded work section above) is expected to be done as teamwork performed by self-organized teams of, generally, up to three students enrolled in this course. Accordingly, a collaborative work within these project teams on matters specific to their respective final project is permitted and required (all members of the project team must contribute). Please note, however, that this permission applies to the above final-project work only; it does **not** apply to any other gradable items (such homework assignments, quizzes/exams).

**Important notice regarding email:**

- Any email you send to me should to be sent
**only**from your UMass Lowell student email account (i.e., “at student uml edu”). I will reply**only**to email messages you send from your UML student email address. **No email attachments:**If you email me, please note that I will NOT accept emails that contain any attachments. In general, there should be no need to include any attachments in your email messages to me. Unlikely exceptions to this rule would be on a one-time, case-by-case basis**AND**would require you to request (with an appropriate justification)**and obtain from me****ahead of time**an explicit one-time permission to send a specific attachment. Any and all email messages you send me, or cc/copy me on, that contain any unexpected attachment will not be considered as having been received, will not be replied to, and**may be deleted and/or unopened.**