AI.899: Geometric Deep Learning
This is a course offered by Sungsoo Ahn @ KAIST, graduate school of AI.
Course Information
Location
Room 9509
5th Floor, Building 9
Kim Jaechul Graduate School of AI, KAIST
85 Hoegi-ro, Dongdaemun-gu
Seoul 02455, Republic of Korea
You can also join via Zoom link.
Schedule
We have the lecture at every Monday/Wednseday 10:30-12:00pm, except for holidays, ICLR, and exam periods.
- I will announce more details on the make-up lectures soon.
| Date | Day | Topic |
|---|---|---|
| 2/24 | Mon | Course Introduction |
| 2/26 | Wed | Preliminary: Equivariance and Group Theory |
| 3/3 | Mon | No Class (Independence Day) |
| 3/5 | Wed | Graph Neural Networks |
| 3/10 | Mon | Graph Neural Networks |
| 3/12 | Wed | Graph Neural Networks |
| 3/17 | Mon | Graph Neural Networks |
| 3/19 | Wed | Geometric GNNs |
| 3/24 | Mon | No class (Buisness Trip) |
| 3/26 | Wed | Geometric GNNs |
| 3/31 | Mon | Geometric GNNs |
| 4/2 | Wed | Geometric GNNs |
| 4/7 | Mon | Geometric GNNs |
| 4/9 | Wed | Geometric GNNs |
| 4/14 | Mon | No Class (Mid-Term Exam) |
| 4/16 | Wed | No Class (Mid-Term Exam) |
| 4/21 | Mon | Preliminary: Diffusion and Flow Matching Models |
| 4/23 | Wed | No Class (ICLR) |
| 4/28 | Mon | No Class (ICLR) |
| 4/30 | Wed | Preliminary: Diffusion and Flow Matching Models |
| 5/5 | Mon | No Class (Buddha’s Birthday & Children’s Day) |
| 5/7 | Wed | Geometric Generative Models |
| 5/12 | Mon | Geometric Generative Models |
| 5/14 | Wed | Geometric Generative Models |
| 5/19 | Mon | Geometric Generative Models |
| 5/21 | Wed | Applications |
| 5/26 | Mon | Invited Lecture (Jinwoo Kim) |
| 5/28 | Wed | Applications |
| 6/2 | Mon | Applications |
| 6/4 | Wed | Applications |
| 6/9 | Mon | No Class (Final Exam) |
| 6/11 | Wed | No Class (Final Exam) |
Blog Post
Topics
- Summarizing and augmenting the course, about the lectures on “geometric GNNs” or “geometric generative models”.
- Reviewing two or more of the recent papers related to the course, must include one of the papers in the predefined list.
- Your own research project related to the course (must get the instructor’s approval beforehand).
Important Deadlines All the deadlines are until 11:59 PM, KST.
- May 25th: Blog posts submission
- June 1st: Review submission
- June 8th: Final blog revision (incorporating reviews)
- June 12th: Reviewer acknowledgement and score adjustment
Blog Topics
- Course content summary:
- Focus on “geometric GNNs” or “geometric generative models”
- Include your own insights and additional research
- Paper reviews:
- Review at least two recent papers
- One must be from the predefined list
- Research project:
- Must be related to course content
- Requires instructor’s prior approval
Peer Review
- Each student will review 5 blog posts
- Additional review details will be provided soon
Submission Instructions
Blog posts (and the abstracts) will be submitted through editing the sample blog (See ./posts/2025-04-28-STUDENTID.md). Adding new type of contents, e.g., interactive figures, is allowed as long as the new contents do not require installing new packages.
- Blog post content:
- File:
./_posts/2025-04-28-STUDENTID.md - Example:
./_posts/2025-04-28-20231234.md
- File:
- Bibliography
- File:
./assets/bibliography/2025-04-28-STUDENTID.bib - Example:
./assets/bibliography/2025-04-28-20231234.bib
- File:
- Blog assets (images, etc.):
- Directory:
./assets/ASSET_TYPE/2025-04-28-STUDENTID/ - Example 1:
./assets/img/2025-04-28-20231234/figure1.png - Example 2:
./assets/video/2025-04-28-20231234/video1.mp4
- Directory: