Graph Classification with GNNs: Optimisation, Representation and Inductive Bias

• URL: http://arxiv.org/abs/2408.09266v2
• Date: Fri, 23 Aug 2024 09:55:08 GMT
• Title: Graph Classification with GNNs: Optimisation, Representation and Inductive Bias
• Authors: P. Krishna Kumar a, Harish G. Ramaswamy,
• Abstract summary: We argue that such equivalence ignores the accompanying optimization issues and does not provide a holistic view of the GNN learning process. We prove theoretically that the message-passing layers in the graph, have a tendency to search for either discriminative subgraphs, or a collection of discriminative nodes dispersed across the graph.
• Score: 0.6445605125467572
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Theoretical studies on the representation power of GNNs have been centered around understanding the equivalence of GNNs, using WL-Tests for detecting graph isomorphism. In this paper, we argue that such equivalence ignores the accompanying optimization issues and does not provide a holistic view of the GNN learning process. We illustrate these gaps between representation and optimization with examples and experiments. We also explore the existence of an implicit inductive bias (e.g. fully connected networks prefer to learn low frequency functions in their input space) in GNNs, in the context of graph classification tasks. We further prove theoretically that the message-passing layers in the graph, have a tendency to search for either discriminative subgraphs, or a collection of discriminative nodes dispersed across the graph, depending on the different global pooling layers used. We empirically verify this bias through experiments over real-world and synthetic datasets. Finally, we show how our work can help in incorporating domain knowledge via attention based architectures, and can evince their capability to discriminate coherent subgraphs.

Related papers

• Exploring Consistency in Graph Representations:from Graph Kernels to Graph Neural Networks [4.235378870514331]
  Graph Networks (GNNs) have emerged as a dominant approach in graph representation learning. We bridge the gap between neural network methods and kernel approaches by enabling GNNs to consistently capture structures in their learned representations. Inspired by these findings, we conjecture that the consistency in the similarities of graph representations across GNN layers is crucial in capturing relational structures and enhancing graph classification performance.
  arXiv  Detail & Related papers  (2024-10-31T09:07:08Z)
• Contextualized Messages Boost Graph Representations [1.5178009359320295]
  This paper investigates the ability of graph networks (GNNs) to process data that may be represented as graphs. It shows that only a few GNNs are investigated across all levels of capability. A mathematical discussion on the relationship between SIRGCN and widely used GNNs is laid out to put the contribution into context.
  arXiv  Detail & Related papers  (2024-03-19T08:05:49Z)
• Generalization Limits of Graph Neural Networks in Identity Effects Learning [12.302336258860116]
  Graph Neural Networks (GNNs) have emerged as a powerful tool for data-driven learning on various graph domains. We establish new generalization properties and fundamental limits of GNNs in the context of learning so-called identity effects. Our study is motivated by the need to understand the capabilities of GNNs when performing simple cognitive tasks.
  arXiv  Detail & Related papers  (2023-06-30T20:56:38Z)
• Representation Power of Graph Neural Networks: Improved Expressivity via Algebraic Analysis [124.97061497512804]
  We show that standard Graph Neural Networks (GNNs) produce more discriminative representations than the Weisfeiler-Lehman (WL) algorithm. We also show that simple convolutional architectures with white inputs, produce equivariant features that count the closed paths in the graph.
  arXiv  Detail & Related papers  (2022-05-19T18:40:25Z)
• Discovering the Representation Bottleneck of Graph Neural Networks from Multi-order Interactions [51.597480162777074]
  Graph neural networks (GNNs) rely on the message passing paradigm to propagate node features and build interactions. Recent works point out that different graph learning tasks require different ranges of interactions between nodes. We study two common graph construction methods in scientific domains, i.e., emphK-nearest neighbor (KNN) graphs and emphfully-connected (FC) graphs.
  arXiv  Detail & Related papers  (2022-05-15T11:38:14Z)
• Graph Neural Networks for Graphs with Heterophily: A Survey [98.45621222357397]
  We provide a comprehensive review of graph neural networks (GNNs) for heterophilic graphs. Specifically, we propose a systematic taxonomy that essentially governs existing heterophilic GNN models. We discuss the correlation between graph heterophily and various graph research domains, aiming to facilitate the development of more effective GNNs.
  arXiv  Detail & Related papers  (2022-02-14T23:07:47Z)
• Fair Node Representation Learning via Adaptive Data Augmentation [9.492903649862761]
  This work theoretically explains the sources of bias in node representations obtained via Graph Neural Networks (GNNs) Building upon the analysis, fairness-aware data augmentation frameworks are developed to reduce the intrinsic bias. Our analysis and proposed schemes can be readily employed to enhance the fairness of various GNN-based learning mechanisms.
  arXiv  Detail & Related papers  (2022-01-21T05:49:15Z)
• Generalizing Graph Neural Networks on Out-Of-Distribution Graphs [51.33152272781324]
  Graph Neural Networks (GNNs) are proposed without considering the distribution shifts between training and testing graphs. In such a setting, GNNs tend to exploit subtle statistical correlations existing in the training set for predictions, even though it is a spurious correlation. We propose a general causal representation framework, called StableGNN, to eliminate the impact of spurious correlations.
  arXiv  Detail & Related papers  (2021-11-20T18:57:18Z)
• Is Homophily a Necessity for Graph Neural Networks? [50.959340355849896]
  Graph neural networks (GNNs) have shown great prowess in learning representations suitable for numerous graph-based machine learning tasks. GNNs are widely believed to work well due to the homophily assumption ("like attracts like"), and fail to generalize to heterophilous graphs where dissimilar nodes connect. Recent works design new architectures to overcome such heterophily-related limitations, citing poor baseline performance and new architecture improvements on a few heterophilous graph benchmark datasets as evidence for this notion. In our experiments, we empirically find that standard graph convolutional networks (GCNs) can actually achieve better performance than
  arXiv  Detail & Related papers  (2021-06-11T02:44:00Z)
• Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting [63.04999833264299]
  "Graph Substructure Networks" (GSN) is a topologically-aware message passing scheme based on substructure encoding. We show that it is strictly more expressive than the Weisfeiler-Leman (WL) graph isomorphism test. We perform an extensive evaluation on graph classification and regression tasks and obtain state-of-the-art results in diverse real-world settings.
  arXiv  Detail & Related papers  (2020-06-16T15:30:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.