Explore the concept of learning by transference in large graphs through this IEEE Signal Processing Society webinar presented by Alejandro Ribeiro from UPenn. Delve into topics such as graph limits, convergence results, transferability, and multirobot consensus. Examine technical aspects including graphic convolutions, graph definitions, frequency representation, and graph neural networks. Gain insights into the demodulation trick, graphone convolution, and graph design. Understand the importance of this subject in the context of data science on graphs and its applications in signal processing.
Overview
Syllabus
Introduction
Why
How
Questions
Graph Limits
Convergence
Results
Transferability
Multirobot Consensus
Technical Part
Graphic Convolutions
Graphs
Definitions
Frequency representation
Review
Transferability Analysis
Graph Neural Networks
Graph Filters vs Graph Neural Networks
Demodulation Trick
Conclusion
Graphone Convolution
Designing Graphs
Taught by
IEEE Signal Processing Society
