arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2608.30152cs.LG

基于逆与碰撞的混合距离-频谱图位置编码节点定位可达性

Converse and Collision-Based Achievability for Node Localization with Hybrid Distance-Spectral Graph Positional Encodings

Zimo Yan, Yifan Li, Hao Li, Zheng Xie, Chang Liu, Zheming Tu, Yuan Wang

首次发表
浏览论文内容

中文总结 AI 辅助

该研究针对图位置编码的节点识别问题,提出混合距离-频谱编码,推导碰撞分解与信息公式,通过随机正则图和拉普拉斯能量坐标验证,实验显示其定位与句法树几何恢复性能优于单一编码基线。

中文摘要 AI 辅助

图位置编码被广泛应用于图神经网络和图Transformer中,但编码本身何时能识别节点仍不明确。本研究提出一种混合距离-频谱编码,将锚点距离轮廓与量化的低频拉普拉斯能量坐标相结合。将该编码视为观测映射,得到单纯形细化的逆、精确碰撞分解κ_H=κ_Dκ_{S|D},以及碰撞信息I_H=-logκ_D-logκ_{S|D}。针对随机正则图,通过有界相关高斯波替代使判据明确;针对实际拉普拉斯能量坐标,给出距离条件频谱碰撞条件,该条件足以实现条件实际坐标可达性。实验表明,I_H/log n可校准定位成功率,在通用依存树的仅位置编码结构任务探测中,混合编码比仅距离或仅频谱基线更能恢复句法树几何结构。

英文摘要

Graph positional encodings are widely used in graph neural networks and graph Transformers, yet it remains unclear when the code itself can identify nodes. We study a hybrid distance-spectral encoding that combines anchor-distance profiles with quantized low-frequency Laplacian-energy coordinates. Treating the encoding as an observation map yields a simplex-refined converse, an exact collision factorization \(κ_H=κ_Dκ_{S|D}\), and the collision information \(I_H=-\logκ_D-\logκ_{S|D}\). On random regular graphs, the criterion is made explicit through a bounded-correlation Gaussian-wave surrogate; for actual Laplacian-energy coordinates, we give the distance-conditioned spectral collision condition sufficient for conditional actual-coordinate achievability. Experiments show that \(I_H/\log n\) calibrates localization success, and PE-only structural task probes on Universal Dependencies trees show that hybrid encodings better recover syntactic-tree geometry than distance-only or spectral-only baselines.

发表机构

  • National University of Defense Technology(国防科技大学)
  • Wuhan University(武汉大学)

机构由 AI 辅助整理,请以论文原文为准。

↑