This course is an introductory course on traditional statistical machine learning. Topics include: linear regression, classification, sampling methods, model selection, tree-based methods, support vector machine, deep learning, unsupervised learning, survival analysis, and hypothesis testing. Prior knowledge in basic statistics and programming (Python) is recommended.
Course Codes: AI5213 / EC4213 / AI4021 / CT5303 / ET5402 / FE5402
Meet our team, led by a professor and supported by three TAs, facilitating your learning experience.
After class / Wed 4:00-5:00pm
AI Building (S7), Room 204
AI Building (S7), Room 208
AI Building (S7), Room 208
Dasan Building (C9) Room 104
- Communication: The course schedule and all resources (e.g. lecture slides, discussion worksheets) will be posted on the course website: https://sundong.kim/courses/mldl23f. All class discussions, announcements and other communication will take place via Ed Discussion. If you need to contact the course staff privately, please make a private question on Ed.
- Access the EC4213/AI5213 Ed Discussion forum. If you haven’t already been added to the class, use this invitation link.
- Submit your assignments at the EC4213/AI5213 Gradescope. If you need the entry code, find it on Ed Discussion in the post entitled “Gradescope invitation link”
Textbook & References
PDFs are available on below links.
- Introduction to Statistical Learning (ISL)
- Elements of Statistical Learning (ESL)
- Probabilistic Machine Learning: An Introduction (PML)
- Video lectures of the authors
Homework and Exams
- HW 1 is due Sunday, Sep 10 at 11:59PM
- HW 2 is due Sunday, Sep 24 at 11:59PM
- HW 3 is due Sunday, Oct 15 at 11:59PM
- HW 4 is due Sunday, Nov 5 at 11:59PM
- HW 5 is due Sunday, Nov 19 at 11:59PM
- HW 6 is due Sunday, Dec 3 at 11:59PM
The midterm will take place on Monday, Oct 23 in GIST College Building C, Room 104.
The final exam will take place on Wednesday, Dec 6 in GIST College Buildge C, Room 104.
If you have a conflict with any of the exams, let us know as soon as possible, and we will schedule a makeup. All exams will be closed book with a single A4-size (both sides) crib sheet. The crib sheets must be handwritten, which is created and printed on an iPad are not acceptable. Exams will cover material from lecture, sections, the readings, and the project. In particular, you are likely to do poorly on the exams and in the course if you do not do your share of the homeworks and projects.
You will be taking part in a machine learning contest, either as a team or individually. Participate in one of the following challenges:
- 데이콘: HD 현대 AI Challenge (Team size: up to 3)
- Forecast vessel waiting times at ports from voyage and operational data.
- Leaderboard closes: Oct 30, 2023
- NeurIPS 2023: CityLearn Challenge Forecast track (Team size: up to 4)
- Predict building energy management load profiles using regression models.
- Leaderboard closes: Oct 31, 2023
- ARCathon 2023 (Team size: up to 4)
- Develop an AI capable of autonomously solving novel tasks.
- Leaderboard closes: Dec 1, 2023
- Kaggle: Child Mind Institute - Detect sleep state (Team size: up to 4)
- Identify sleep onset and wake states using wrist accelerometer data.
- Leaderboard closes: Dec 5, 2023
Upon the conclusion of your chosen contest, your project outcomes will be showcased through two mediums: a poster presentation and a detailed project report.
The purpose of the poster presentation is to share your project outcomes with your peers. It’s a chance for you to showcase the hard work you’ve put into your project and engage in insightful discussions with your colleagues.
- Poster Specifications: A0 size (841 x 1189 mm), portrait orientation (세로 방향), here are some templates: A, B, C.
- Printing Information: For printing, reach out to 첨단문화사. To ensure timely printing, ask them to print your posters at least 4-5 days in advance.
- Address: LG Library A, 1F, GIST
- Direct Line: 062-973-2686
- E-mail: firstname.lastname@example.org
- Venue: Bridge between GIST College B and C (N5 and N6, 3F)
- Presentation: Dec 4, 13:00-14:30 (TAs will be available from 12:00 to help you set up your posters)
- Workload Adjustment: To address concerns about additional workload, HW6 will be substituted by this poster presentation. Submit the PDF version of your poster through GradeScope by Dec 3, Sunday 23:59.
The posters will be grouped by project, and you’ll be informed of your specific placement location. Let’s make it a great event!
Articulating your ideas and experimental results in writing is pivotal, not only for this course but for your future academic and research endeavors. The project report complements your poster presentation, providing a comprehensive insight into your analytical capabilities and methodological rigor.
- Report deadline: Dec 16 (Sat), 23:59.
- The final report should be written in English using the following template.
- Example paper (Jeong et al., 2023) - feel free to copy this project instead.
- Use KDD Cup 2023 papers as a guide for your report structure.
- Maximum report length is 8 pages; no minimum length.
- Include a snapshot showing your team’s rank on the leaderboard.
- Submit via Gradescope by one team member.
- Report should list full names, student IDs, and emails of all team members.
- If your approach has academic merit, consider submitting to international conferences like KDD 2024, SIGIR 2024 or ICML 2024 after the semester ends.
The policy is simple: there are no slip dates. If assignments are late, they are increasingly penalized as follows: within 24 hours, you lose 10%; within 48 hours, you lose 20%; within 72 hours, you lose 40%. More than three days late, you can no longer hand-in the assignment. Note that the penalty scheme applies to project deadlines too.
The course will be graded on a curve (25% A, 25% B+, 25% B, 15% C+, 10% C). Median course grade is generally placed in neighborhood of B/B+ border. A+ is reserved for the very few best students in the class. The curve can shift up for an especially excellent class, as indicated but strong classroom interaction and outstanding project implementations. Graduate students and reentry students are not included in establishing the curve (to be fairer to undergraduates), but they will receive grades based on where they would fall on the curve.
- Six homework (48%, 8% each)
- Midterm (16%)
- Final Exam (16%)
- Project: ML Challenge (20%)
Collaboration Policy & Honor Code
Study groups are allowed. It is also OK to get clarification (but not solutions) from books or online resources, again after you have thought about the problems on your own. However, we expect students to understand and complete their own assignments. Each student must write down the solution independently and hand in one assignment per student, which means you write your solution after closing the book and all your notes, without helped by your colleagues. If you studied together as a group, please cite your collaborators fully and completely at the top of your assignment (e.g., “Junho explained to me what is asked in Question 2.1”). When in doubt about collaboration details, please ask us on Ed discussion.
If elements of two assignments are determined to be clearly very similar, we believe that they were done together or one was copied from the other), then the course grade for all students involved in the incident will be reduced by one letter grade for the first offense, and to an F for the second offense. (All means both the copy-ers and the copy-ees). The grade for that assignment will also be reduced to 0. More serious cases of cheating (e.g., cheating on exams) will lead to severe consequences ranging from a grade of “F” on the class to suspension from the University.