This course covers the learning outcomes and goals of understanding machine learning with graphs. Students will learn about the structure of graphs, properties of networks, community structure, spectral clustering, graph representation learning, graph neural networks, deep generative models for graphs, link analysis, network effects, influence maximization, outbreak detection, network evolution, reasoning over knowledge graphs, limitations of graph neural networks, and applications of graph neural networks. The course teaches skills such as message passing, node classification, implementation with Pytorch Geometric, and probabilistic contagion modeling. The teaching method includes lectures on various topics related to machine learning with graphs. The intended audience for this course is individuals interested in machine learning, graph analysis, and neural networks.
Overview
Syllabus
Lecture 1 Introduction; Structure of Graphs.
Lecture 2 Properties of Networks And Random Graph Models.
Lecture 3 Motifs and Structural Roles in Networks.
Lecture 4 Community Structure in Networks.
Lecture 5 Spectral Clustering.
Lecture 6 Message Passing and Node Classification.
Lecture 7 Graph Representation Learning.
Lecture 8 Graph Neural Networks.
Lecture 9 Graph Neural Networks Implementation with Pytorch Geometric.
Lecture 10 Deep Generative Models for Graphs.
Lecture 11 Link Analysis - PageRank.
Lecture 12 Network Effects and Cascading Behavior.
Lecture 13 Probabilistic Contagion and Models of Influence.
Lecture 14 Influence Maximization in Networks.
Lecture 15 Outbreak Detection in Networks.
Lecture 16 Network Evolution.
Lecture 17 Reasoning over Knowledge Graphs.
Lecture 18 Limitations of Graph Neural Networks.
Lecture 19 Applications of Graph Neural Networks.
Taught by
Hussain Kara Fallah
