Graphs have become increasingly popular in modern data processing tasks, especially large-scale data analysis. Traditionally a tool from the viewpoints of mathematics and computer science, it starts attracting more and more attention from the signal processing community. My research tries to fill the ever-decreasing gap between signal processing and machine learning, by focusing on graph-based signal processing and learning algorithms, with emerging applications to mobile and online social network analysis.

More specifically, I have been working on the following research topics during my PhD studies:

– Clustering with multi-layer graphs
– Learning dictionaries for graph signals
– Classification of graph signals using graph-based transforms
– Learning graphs from signal observations