利用未标记节点预测优化图分类，无需改架构
Graph Transductive Sharpening: Leveraging Unlabeled Predictions in Node Classification
- 基于预测置信度提取未标记节点训练信号
- 在多个基准上提升分类准确率，无需修改模型结构
- 适合希望不改架构提升性能的研究者
在全图可观测但标签仅部分可用的归纳设置中，半监督节点分类进展主要集中在架构创新。本文另辟蹊径，聚焦训练目标。观察到归纳模型在训练时对所有节点（包括无标签节点）均有预测输出，这些未标记节点的预测可能蕴含有效训练信号，但传统监督目标因缺乏真实标签而忽略。受交叉熵分解为依赖标签的对齐项与独立标签的熵项启发，我们提出以预测置信度作为无标签情况下的信号提取方式。由此提出归纳锐化（Transductive Sharpening, TS）：一种损失层面的改进，在未标记节点上最小化预测熵，同时在标记节点上进行补偿。在广泛节点分类基准上评估显示，该方法在不改变主干架构的前提下实现一致性能提升。代码已公开于 https://github.com/transductive-sharpening/tunedGNN。
原文摘要 · Abstract (English)
In the transductive setting, where the full graph is observed but node labels are only partially available, progress in semi-supervised node classification has largely focused on architectural innovation. In this paper, we revisit an orthogonal axis: the training objective. We start from a simple observation: transductive models produce predictions for every node during training, including nodes without labels. These unlabeled-node predictions may contain useful training signal, but standard supervised objectives discard them because no ground-truth labels are available. Inspired by the decomposition of cross-entropy into a label-dependent alignment term and a label-independent entropy term, we propose prediction confidence as a natural way to extract this signal in the absence of labels. This motivates Transductive Sharpening (TS): a loss-level modification that minimizes prediction entropy on unlabeled nodes while counterbalancing this effect on labeled nodes. We evaluate Transductive Sharpening across a wide range of node-classification benchmarks and observe consistent performance improvements without requiring any changes to the backbone architecture. Code is available at https://github.com/transductive-sharpening/tunedGNN.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。