Graph Few-Shot Learning via Adaptive Spectrum Experts and Cross-Set Distribution Calibration

2025-10-14

AI Summary

Existing graph few-shot learning methods face two key challenges: (1) reliance on predefined global graph filters, limiting adaptability to local topological heterogeneity; and (2) sparse support-set samples, causing distributional shift between support and query sets and impairing generalization. To address these, we propose AdaptSpectra—a meta-learning framework integrating adaptive spectral experts with cross-set distribution calibration. Its core innovations are: (i) learnable local spectral experts that dynamically generate node-neighborhood-specific graph filters; and (ii) a cross-set feature distribution calibration module that explicitly aligns latent-space distributions of support and query sets. Evaluated on multiple few-shot graph classification benchmarks, AdaptSpectra consistently outperforms state-of-the-art methods, demonstrating superior capability in modeling local structural patterns and robustness to distributional shifts.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Local Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning 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

Graph few-shot learning has attracted increasing attention due to its ability to rapidly adapt models to new tasks with only limited labeled nodes. Despite the remarkable progress made by existing graph few-shot learning methods, several key limitations remain. First, most current approaches rely on predefined and unified graph filters (e.g., low-pass or high-pass filters) to globally enhance or suppress node frequency signals. Such fixed spectral operations fail to account for the heterogeneity of local topological structures inherent in real-world graphs. Moreover, these methods often assume that the support and query sets are drawn from the same distribution. However, under few-shot conditions, the limited labeled data in the support set may not sufficiently capture the complex distribution of the query set, leading to suboptimal generalization. To address these challenges, we propose GRACE, a novel Graph few-shot leaRning framework that integrates Adaptive spectrum experts with Cross-sEt distribution calibration techniques. Theoretically, the proposed approach enhances model generalization by adapting to both local structural variations and cross-set distribution calibration. Empirically, GRACE consistently outperforms state-of-the-art baselines across a wide range of experimental settings. Our code can be found here.

Problem

• Adapting graph filters to local topological heterogeneity
• Calibrating distribution gaps between support and query sets
• Improving generalization in graph few-shot learning scenarios

Innovation

• Adaptive spectrum experts handle local graph heterogeneity
• Cross-set distribution calibration aligns support-query distributions
• Combined approach enhances generalization in few-shot learning