Uncertainty Quantification on Graph Learning: A Survey

2024-04-23

arXiv.org

AI Summary

Graphical models (PGMs) and graph neural networks (GNNs) suffer from performance degradation and diminished reliability due to uncertainties arising from data randomness and modeling complexity, necessitating dedicated uncertainty quantification (UQ) methods. This work presents the first systematic survey of UQ in graph learning, innovatively decoupling existing approaches into two orthogonal axes: *uncertainty representation* and *uncertainty handling*. We unify and analyze the adaptation mechanisms of key UQ techniques—including Bayesian inference, Monte Carlo Dropout, ensemble methods, confidence calibration, and information-theoretic entropy—within both PGMs and GNNs. Building upon this analysis, we propose the first comprehensive UQ methodology framework tailored specifically for graph-structured data. The survey explicitly identifies critical challenges, such as structural dependency, message-passing bias, and the lack of standardized evaluation metrics, while highlighting promising research directions. This review serves as an authoritative reference for advancing both theoretical foundations and real-world deployment of trustworthy graph AI.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Graphical Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Graphical models have demonstrated their exceptional capabilities across numerous applications, such as social networks, citation networks, and online recommendation systems. Despite these successes, their performance, confidence, and trustworthiness are often limited by the inherent randomness of data in nature and the challenges of accurately capturing and modeling real-world complexities. This has increased interest in developing uncertainty quantification (UQ) techniques tailored to graphical models. In this survey, we comprehensively examine these existing works on UQ in graphical models, focusing on key aspects such as foundational knowledge, sources, representation, handling, and measurement of uncertainty. This survey distinguishes itself from most existing UQ surveys by specifically concentrating on UQ in graphical models, particularly probabilistic graphical models (PGMs) and graph neural networks (GNNs). We elaborately categorize recent work into two primary areas: uncertainty representation and uncertainty handling. By offering a comprehensive overview of the current landscape, including both established methodologies and emerging trends, we aim to bridge gaps in understanding and highlight key challenges and opportunities in the field. Through in-depth discussion of existing works and promising directions for future research, we believe this survey serves as a valuable resource for researchers, inspiring them to cope with uncertainty issues in both academic research and real-world applications.

Problem

• Quantifying uncertainty in graph learning models
• Surveying UQ techniques for probabilistic and neural graphical models
• Addressing uncertainty sources, representation, and handling in graphs

Innovation

• Survey focuses on UQ in graphical models
• Covers uncertainty representation and handling
• Includes PGMs and GNNs for UQ