- Instructor: William L. Hamilton
- Term: Winter 2019
- When: Tuesdays and Thursdays, 2:35-3:55pm
- Where: Adams Auditorium
- See the syllabus and schedule for more details.
- Head TA: Joey Bose - email@example.com
- Aishik Chakraborty - firstname.lastname@example.org
- Komal Teru - email@example.com
- Jin Dong - firstname.lastname@example.org
- Xin Tong Wang - email@example.com
- Zhilong Chen - firstname.lastname@example.org
- Etienne Dennis - email@example.com
- Amy Ruskin - firstname.lastname@example.org
- Gandharv Patil - email@example.com
- Haque Ishfaq - firstname.lastname@example.org
The course will cover selected topics and new developments in data mining and applied machine learning, with a particular emphasis on good methods and practices for effective deployment of real systems. We will study commonly used algorithms and techniques, including linear and logistic regression, clustering, neural networks, support vector machines, decision trees and more. We will also discuss methods to address practical issues such as empirical validation, feature selection, dimensionality reduction, and error estimation.
Important note: Students who took COMP-652 in 2013 or before CANNOT take COMP-551. Students who took COMP-652 in Winter 2014 or after (or intend to take it) can take COMP-551. Contents of both courses have been designed to avoid too much overlap. COMP-551 focuses on the practical application of machine learning, whereas COMP-652 (starting in Winter 2014) focuses on theoretical analysis of machine learning, reinforcement learning, bandits and analysis of time series.
TA Contact Info / Office Hours:
Please add COMP 551 to the subject for all emails.
There are office hours every week, but each individual TA holds their office hours every other week. The first week of class (Jan 7-11) is considered an Odd week, so TA’s holding office hours on Odd weeks will be available on weeks 1,3,5 etc … \
- Joey Bose: Tuesday 10:30 - 12:30 pm in McConnel 104 (Even Weeks)
- Amy Ruskin: Friday 12 - 2 pm, in Trottier 3110 (Even Weeks)
- Xin Tong (Alex) Wang: Thursday 9:45 - 11:45 am, in Trottier 3104 (Odd Weeks)
- Jin Dong: Thursday 4:10 - 6:10 pm in Trottier 3104 (Even Weeks)
- Komal Teru: Thursday 4:15 - 6:15 pm in McConnel 104 (Odd Weeks)
- Etienne Denis: Friday 2 - 4 pm in Trottier 3110 (Even Weeks)
- Aishik Chakraborty: Monday 3 - 5 pm, McConnel 111 (Odd Weeks)
- Zhilong Chen: Wednesday 11 - 1 pm in McConnel 229 (Even Weeks)
- Gandharv Patil: Friday 2 - 4 pm in McConnel 436 (Odd Weeks)
- Haque Ishfaq: Friday 1:30 - 3:30 pm in McConnel 107 (Odd Weeks)
List of topics (subject to minor changes)
- Linear regression.
- Linear classification.
- Performance evaluation, overfitting, cross-validation, bias-variance analysis, and error estimation.
- Naive Bayes.
- Decision trees, regression trees, and ensemble methods.
- Support vector machines.
- Artificial neural networks and deep learning (e.g., RNNs and CNNs).
- Feature selection and dimensionality reduction.
- Unsupervised learning and clustering.
- Semi-supervised learning.
- Generative models.