This is the course homepage for CPSC 330: Applied Machine Learning at the University of British Columbia. You are looking at the current version (Sep-Dec 2022). You can find the earlier versions here:
- Sep-Dec 2020 taught by Mike Gelbart
- Sep-Dec 2021 taught by Varada Kolhatkar
- Jan-April 2022 taught by Giulia Toti
- May-June 2022 taught by Mehrdad Oveisi
Instructor: Varada Kolhatkar
© 2022 Varada Kolhatkar and Mike Gelbart
Software licensed under the MIT License, non-software content licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License. See the license file for more information.
- Calendar
- Course GitHub page
- Course Jupyter book
- iClicker Cloud
- Piazza
- Canvas: You will find Gradescope, Piazza, and Panopto links in Canvas
- Course videos YouTube channel
- Syllabus / administrative info
- Other course documents
Usually the homework assignments will be due on Mondays (except next week) and will be released on Tuesdays. I'll also add the due dates in the Calendar. If you find inconsistencies in due dates, follow the due date in the Calendar. For this course, we'll assume that the Calendar is always right!
| Assessment | Due date | Where to find? | Where to submit? |
|---|---|---|---|
| hw1 | Sept 13, 11:59pm | Github repo | Gradescope |
| Syllabus quiz | Sept 19, 11:59pm | Canvas | Canvas |
| hw2 | Sept 19, 11:59pm | Github repo | Gradescope |
| hw3 | Oct 03, 11:59pm | Github repo | Gradescope |
| hw4 | Oct 11, 11:59pm | Github repo | Gradescope |
| Midterm | Oct 27, during class time | Canvas | [Canvas] |
| hw5 | Oct 31, 11:59pm | Github repo | Gradescope |
| (https://canvas.ubc.ca/courses/101888) | |||
| hw6 | November 10th | Github repo | Gradescope |
| hw7 | November 22nd | Github repo | Gradescope |
| hw8 | November 29th | Github repo | Gradescope |
| hw9 | December 6th | Github repo | Gradescope |
| Final exam | December 15th | Canvas | Canvas |
Live lectures: The lectures will be in-person in DMP 310 from 11am to 12:20pm, as marked in the Calendar. The lectures will be live streamed. You can find the link of Panopto videos in Canvas. That said, consider the recordings a backup resource and do not completely rely on it. You will get a lot more out of the course if you show up in person.
This course will be run in a semi flipped classroom format. There will be pre-watch videos for many lectures, at least in the first half of the course. All the videos are available on YouTube and are posted in the schedule below. Try to watch the assigned videos before the corresponding lecture. During the lecture, I'll summarize the important points from the videos and focus on demos, iClickers, and Q&A.
I'll be developing lecture notes directly in this repository. So if you check them before the lecture, they might be in a draft form. Once they are finalized, I'll post them in the Course Jupyter book.
| Date | Topic | Assigned videos | vs. CPSC 340 |
|---|---|---|---|
| Sep 6 | UBC Imagine Day - no class | ||
| Sep 8 | Course intro | 📹 Pre-watch: 1.0 | n/a |
| Sep 13 | Decision trees | 📹 Pre-watch: 2.1, 2.2, 2.3, 2.4 | less depth |
| Sep 15 | ML fundamentals | 📹 Pre-watch: 3.1, 3.2, 3.3, 3.4 | similar |
| Sep 20 | $k$-NNs and SVM with RBF kernel | 📹 Pre-watch: 4.1, 4.2, 4.3, 4.4 | less depth |
| Sep 22 | Preprocessing, sklearn pipelines |
📹 Pre-watch: 5.1, 5.2, 5.3, 5.4 | more depth |
| Sep 27 | More preprocessing, sklearn ColumnTransformer, text features |
📹 Pre-watch: 6.1, 6.2 | more depth |
| Sep 29 | Linear models | 📹 Pre-watch: 7.1, 7.2, 7.3 | less depth |
| Oct 04 | Hyperparameter optimization, overfitting the validation set | 📹 Pre-watch: 8.1, 8.2 | different |
| Oct 06 | Evaluation metrics for classification | 📹 Reference: 9.2, 9.3,9.4 | more depth |
| Oct 11 | Regression metrics | 📹 Pre-watch: 10.1 | more depth on metrics less depth on regression |
| Oct 13 | Ensembles | 📹 Pre-watch: 11.1, 11.2 | similar |
| Oct 18 | Feature importances, model interpretation | 📹 Pre-watch: 12.1,12.2 | feature importances is new, feature engineering is new |
| Oct 20 | Feature engineering and feature selection | None | less depth |
| Oct 25 | Midterm review | ||
| Oct 27 | Midterm | ||
| Nov 1 | Clustering | 📹 Pre-watch: 14.1, 14.2, 14.3 | less depth |
| Nov 3 | More clustering | 📹 Pre-watch: 15.1, 15.2, 15.3 | less depth |
| Nov 8 | Simple recommender systems | less depth | |
| Nov 10 | Midterm break - no class | ||
| Nov 15 | Text data, embeddings, topic modeling | 📹 Pre-watch: 16.1, 16.2 | new |
| Nov 17 | Neural networks and computer vision | less depth | |
| Nov 22 | Time series data | (Optional) Humour: The Problem with Time & Timezones | new |
| Nov 24 | Survival analysis | 📹 (Optional but highly recommended)Calling Bullshit 4.1: Right Censoring | new |
| Nov 29 | Ethics | 📹 (Optional but highly recommended) |
new |
| Dec 1 | Communication | 📹 (Optional but highly recommended) |
new |
| Dec 6 | Model deployment and conclusion | new |
Please read Covid Campus Rules.
Masks: This class is going to be in person. UBC no longer requires students, faculty and staff to wear non-medical masks, but continues to recommend that masks be worn in indoor public spaces.
Your personal health: If you are ill or believe you have COVID-19 symptoms or been exposed to SARS-CoV-2 use the Thrive Health self-assessment tool for guidance, or download the BC COVID-19 Support App for iOS or Android device and follow the instructions provided. Follow the advice from Public Health.
Stay home if you have recently tested positive for COVID-19 or are required to quarantine. You can check this website to find out if you should self-isolate or self-monitor.
Your precautions will help reduce risk and keep everyone safer. In this class, the marking scheme is intended to provide flexibility so that you can prioritize your health and still be able to succeed:
- All course notes will be provided online.
- All homework assignments can be done and handed in online.
- All exams will be held online. (But you need to be present in the classroom to write the exam unless there is a legitimate reason for not doing so.)
- Most of the class activity will be video recorded and will be made available to you.
- There will be at least a few office hours which will be held online.
We are working on this course during the global pandemic. In general, everyone is struggling to some extent. If you tell me you are having trouble, I am not going to judge you or think less of you. I hope you will extend me the same grace!
Here are some ground rules:
- If you are unable to submit a deliverable on time, please reach out before the deliverable is due.
- If you need extra support, the teaching team is here to work with you. Our goal is to help each of you succeed in the course.
- If you are struggling with the material, the new hybrid teaching format, or anything else, please reach out. I will try to find time and listen to you empathetically.
- If I am unable to help you, I might know someone who can. UBC has some great student support resources.
UBC’s Point Grey Campus is located on the traditional, ancestral, and unceded territory of the xwməθkwəy̓əm (Musqueam) peple. The land it is situated on has always been a place of learning for the Musqueam people, who for millennia have passed on their culture, history, and traditions from one generation to the next on this site.
It is important that this recognition of Musqueam territory and our relationship with the Musqueam people does not appear as just a formality. Take a moment to appreciate the meaning behind the words we use:
TRADITIONAL recognizes lands traditionally used and/or occupied by the Musqueam people or other First Nations in other parts of the country.
ANCESTRAL recognizes land that is handed down from generation to generation.
UNCEDED refers to land that was not turned over to the Crown (government) by a treaty or other agreement.
As you begin your journey at UBC, take some time to learn about the history of this land and to honour its original inhabitants.