Stanford CS224W: Machine Learning with Graphs - Stanford CS224W: ML with Graphs | 2021 | Lecture 5.1 - Message passing and Node Classification

For more information about Stanford’s Artificial Intelligence professional and graduate programs, visit: https://stanford.io/3jHRiGj Jure Leskovec Computer Science, PhD From previous lectures, we learn the use of graph representation learning for node classification. In this lecture, we will talk about an alternative approach, message passing. We will introduce the semi-supervised learning on predicting node labels by leveraging correlations that exist in the network. One key concept is the collective classification, which involves three steps including the local classifier that assigns initial labels, the relational classifier that captures correlations, and the collective inference that propagates correlations. To follow along with the course schedule and syllabus, visit: http://web.stanford.edu/class/cs224w/