NetInfoF Framework: Measuring and Exploiting Network Usable Information

Meng-Chieh Lee, Haiyang Yu, Jian Zhang, Vassilis N. Ioannidis, Xiang song, Soji Adeshina, Da Zheng, Christos Faloutsos

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Graph Neural Networks, Information Theory, Heterophily Graphs

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We propose a framework that measures the usable information of graphs and exploits it to solve the graph tasks, namely link prediction and node classification.

Abstract:

Given a node-attributed graph, and a graph task (link prediction or node classification), can we tell if a graph neural network (GNN) will perform well? More specifically, do the graph structure and the node features carry enough usable information for the task? Our goals are
(1) to develop a fast tool to measure how much information is in the graph structure and in the node features, and
(2) to exploit the information to solve the task, if there is enough.
We propose NetInfoF, a framework including NetInfoF_Probe and NetInfoF_Act, for the measurement and the exploitation of network usable information (NUI), respectively. Given a graph data, NetInfoF_Probe measures NUI without any model training, and NetInfoF_Act solves link prediction and node classification, while two modules share the same backbone.
In summary, NetInfoF has following notable advantages:
(a) General, handling both link prediction and node classification;
(b) Principled, with theoretical guarantee and closed-form solution;
(c) Effective, thanks to the proposed adjustment to node similarity;
(d) Scalable, scaling linearly with the input size.
In our carefully designed synthetic datasets, NetInfoF correctly identifies the ground truth of NUI and is the only method being robust to all graph scenarios. Applied on real-world datasets, NetInfoF wins in 11 out of 12 times on link prediction compared to general GNN baselines.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Primary Area: learning on graphs and other geometries & topologies

Submission Number: 959