Affiliation:
1. Pontificia Universidad Catolica de Chile
2. University of Oxford
3. IMFD Chile
4. University of Chile
5. DCC, UChile
Abstract
In this paper we survey our recent results characterizing various graph neural network (GNN) architectures in terms of their ability to classify nodes over graphs, for classifiers based on unary logical formulas- or queries. We focus on the language FOC2, a well-studied fragment of FO. This choice is motivated by the fact that FOC2 is related to theWeisfeiler-Lehman (WL) test for checking graph isomorphism, which has the same ability as GNNs for distinguishing nodes on graphs. We unveil the exact relationship between FOC2 and GNNs in terms of node classification. To tackle this problem, we start by studying a popular basic class of GNNs, which we call AC-GNNs, in which the features of each node in a graph are updated, in successive layers, according only to the features of its neighbors. We prove that the unary FOC2 formulas that can be captured by an AC-GNN are exactly those that can be expressed in its guarded fragment, which in turn corresponds to graded modal logic. This result implies in particular that ACGNNs are too weak to capture all FOC2 formulas. We then seek for what needs to be added to AC-GNNs for capturing all FOC2. We show that it suffices to add readouts layers, which allow updating the node features not only in terms of its neighbors, but also in terms of a global attribute vector. We call GNNs with readouts ACR-GNNs. We also describe experiments that validate our findings by showing that, on synthetic data conforming to FOC2 but not to graded modal logic, AC-GNNs struggle to fit in while ACR-GNNs can generalise even to graphs of sizes not seen during training.
Publisher
Association for Computing Machinery (ACM)
Subject
Information Systems,Software
Reference36 articles.
1. Approximation ratios of graph neural networks for combinatorial problems. Approximation ratios of graph neural networks for combinatorial problems.
2. 13 Description logic
3. Querying graph databases
Cited by
8 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Learning Graph Neural Networks using Exact Compression;Proceedings of the 6th Joint Workshop on Graph Data Management Experiences & Systems (GRADES) and Network Data Analytics (NDA);2023-06-18
2. A Query Language Perspective on Graph Learning;Proceedings of the 42nd ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems;2023-06-18
3. EARLY: Efficient and Reliable Graph Neural Network for Dynamic Graphs;Proceedings of the ACM on Management of Data;2023-06-13
4. Payment behavior prediction on shared parking lots with TR-GCN;The VLDB Journal;2022-01-09
5. On the Expressive Power of Message-Passing Neural Networks as Global Feature Map Transformers;Lecture Notes in Computer Science;2022