Fully-inductive Node Classification on Arbitrary Graphs

2024-05-30

AI Summary

This paper addresses fully inductive node classification in graph machine learning—where models must generalize to entirely unseen graphs with novel nodes, edges, features, and label classes, without access to target-graph priors or fine-tuning. We formally introduce the *fully inductive* setting and propose GraphAny, a framework that formulates inference as the analytical solution of a LinearGNN and introduces a learnable, entropy-normalized distance-driven inductive attention mechanism to enable cross-graph generalization. Crucially, GraphAny performs parameter-free inference and is end-to-end trainable. Trained solely on 120 labeled nodes from the Wisconsin dataset, GraphAny achieves an average accuracy of 67.26% across 30 heterogeneous unseen graphs—outperforming all existing inductive baselines and single-graph transductive methods. This demonstrates both the feasibility and effectiveness of the fully inductive paradigm.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

One fundamental challenge in graph machine learning is generalizing to new graphs. Many existing methods following the inductive setup can generalize to test graphs with new structures, but assuming the feature and label spaces remain the same as the training ones. This paper introduces the fully-inductive setup, where models should perform inference on arbitrary test graphs with new structures, feature and label spaces. We propose GraphAny as the first attempt at this challenging setup. GraphAny models inference on a new graph as an analytical solution to a LinearGNN, which can be naturally applied to graphs with any feature and label spaces. To further build a stronger model with learning capacity, we fuse multiple LinearGNN predictions with learned inductive attention scores. Specifically, the attention module is carefully parameterized as a function of the entropy-normalized distance features between pairs of LinearGNN predictions to ensure generalization to new graphs. Empirically, GraphAny trained on a single Wisconsin dataset with only 120 labeled nodes can generalize to 30 new graphs with an average accuracy of 67.26%, surpassing not only all inductive baselines, but also strong transductive methods trained separately on each of the 30 test graphs.

Problem

• Generalizing to new graphs
• Handling new feature and label spaces
• Improving node classification accuracy

Innovation

• Fully-inductive node classification
• LinearGNN analytical solution
• Entropy-normalized attention scores