Principle of Relevant Information for Graph Sparsification

05/31/2022
by   Shujian Yu, et al.
0

Graph sparsification aims to reduce the number of edges of a graph while maintaining its structural properties. In this paper, we propose the first general and effective information-theoretic formulation of graph sparsification, by taking inspiration from the Principle of Relevant Information (PRI). To this end, we extend the PRI from a standard scalar random variable setting to structured data (i.e., graphs). Our Graph-PRI objective is achieved by operating on the graph Laplacian, made possible by expressing the graph Laplacian of a subgraph in terms of a sparse edge selection vector 𝐰. We provide both theoretical and empirical justifications on the validity of our Graph-PRI approach. We also analyze its analytical solutions in a few special cases. We finally present three representative real-world applications, namely graph sparsification, graph regularized multi-task learning, and medical imaging-derived brain network classification, to demonstrate the effectiveness, the versatility and the enhanced interpretability of our approach over prevalent sparsification techniques. Code of Graph-PRI is available at https://github.com/SJYuCNEL/PRI-Graphs

READ FULL TEXT
research
02/22/2022

Structured Multi-task Learning for Molecular Property Prediction

Multi-task learning for molecular property prediction is becoming increa...
research
05/22/2023

A Fractional Graph Laplacian Approach to Oversmoothing

Graph neural networks (GNNs) have shown state-of-the-art performances in...
research
12/20/2019

Laplacian Flow Dynamics on Geometric Graphs for Anatomical Modeling of Cerebrovascular Networks

Generating computational anatomical models of cerebrovascular networks i...
research
06/26/2023

Interpretable Sparsification of Brain Graphs: Better Practices and Effective Designs for Graph Neural Networks

Brain graphs, which model the structural and functional relationships be...
research
11/11/2018

Analysis vs Synthesis - An Investigation of (Co)sparse Signal Models on Graphs

In this work, we present a theoretical study of signals with sparse repr...
research
05/31/2023

Improving Expressivity of GNNs with Subgraph-specific Factor Embedded Normalization

Graph Neural Networks (GNNs) have emerged as a powerful category of lear...
research
04/06/2023

Inductive Graph Unlearning

As a way to implement the "right to be forgotten" in machine learning, m...

Please sign up or login with your details

Forgot password? Click here to reset