The Final Layer Holds the Key: A Unified and Efficient GNN Calibration Framework

TL;DR

This paper introduces a unified, efficient GNN calibration framework that improves confidence estimates by focusing on class-centroid and node-level calibration, reducing computational overhead and enhancing reliability.

Contribution

It provides a theoretical framework linking confidence calibration to class-centroid and node-level adjustments, proposing a simple method to improve GNN calibration without extra components.

Findings

Reduces GNN under-confidence by lowering weight decay in final layer

Node-level calibration improves confidence at a finer granularity

Method outperforms existing calibration techniques in experiments

Abstract

Graph Neural Networks (GNNs) have demonstrated remarkable effectiveness on graph-based tasks. However, their predictive confidence is often miscalibrated, typically exhibiting under-confidence, which harms the reliability of their decisions. Existing calibration methods for GNNs normally introduce additional calibration components, which fail to capture the intrinsic relationship between the model and the prediction confidence, resulting in limited theoretical guarantees and increased computational overhead. To address this issue, we propose a simple yet efficient graph calibration method. We establish a unified theoretical framework revealing that model confidence is jointly governed by class-centroid-level and node-level calibration at the final layer. Based on this insight, we theoretically show that reducing the weight decay of the final-layer parameters alleviates GNN under-confidence by acting on the class-centroid level, while node-level calibration acts as a finer-grained complement to class-centroid level calibration, which encourages each test node to be closer to its predicted class centroid at the final-layer representations. Extensive experiments validate the superiority of our method.