Course Description

This is a graduate-level research-oriented course offered in Winter 2021. The course aims to introduce and discuss recent advances in graph neural networks (GNNs), with the goal to design deep learning algorithms for graph data for different graph applications. The course contains lecture time by the instructor covering basics of graph neural networks, and paper reading and presentation by students covering recent GNN papers. The students are expected to conduct a team-based research project related to GNN and present the project to the whole class.

Course Requirement

In this course, each student will 1) present one GNN-related paper; 2) finish the course project with group.

This Github repo is the place where each student submit their report and code for both Paper Presentation and Course Project.

For Paper Presentation, the presenter should upload the following files to each given folder before the course:

• Report: briefly summarize the paper. It should be structured as an memo, including introduction, related work, solution, experimental results, major conclusions, pros and cons.
• Demo: an interactive Jupyter Notebook / Colab to show visualize the result of the paper. For most paper, it's highly recommended that you download the source code of the paper, running the main model and see the results or visualization in this notebook.
• Slides: which you use to present at the course. It should contain both the contents of the report and some key results of the code demo. Students are encourage to explore and introduce the background and related work to make the presentation self-contained.
• Quiz: 3 questions about this paper. Use xxx to generate the quiz link, provide automatic gradings and feedback.

Every audience should participate in Q&A after the presentation, and finish the quiz using the quiz link after the course.

For Course Project, each group can firstly create their own github code repo, and then create a subfolder under Course_Project, submit their slides and Github code link under it.

Schedule

• Part I (1/5, 1/7, 1/12): 1Intro.pdf, 2Graph_basics.pdf
• Part II (1/14 - 2/14): Paper Presentation Schedule
• Part III (3/9, 3/11): Course Project Schedule