Limitations of Graph Neural Networks
- To follow along with the course schedule and syllabus, visit: http://web.stanford.edu/class/cs224w/
One essential task to consider before we conduct machine learning on graphs is to find an appropriate way to represent the graphs. What are the factors that will affect our choices as to the representations? In this video, we’ll be looking at the different approaches to abstracting graphs: directed vs. undirected, weighted vs. unweighted, homogeneous vs bipartite, and so on.