发表机构
Center for AI Research, VinUniversity; University of Florida; Posts and Telecommunications Institute of Technology; University of Illinois Chicago（文大人工智能研究中心; 佛罗里达大学; 邮电技术学院; 伊利诺伊大学芝加哥分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
HARMONIA提出神经基元混合与稀疏RRWP聚合，实现可解释且可扩展的图加性模型，兼顾特征特化与结构分解，在解释恢复和预测性能上优于基线。
AI 中文摘要
现有的可解释图加性模型在计算可扩展性或建模灵活性方面仍存在局限。在结构建模方面，以往方法要么面临二次方缩放成本，要么牺牲显式的源到目标贡献分解。在特征组件方面，它们要么依赖逐特征神经网络，要么依赖单一共享基元且特征特化能力有限。我们通过引入HARMONIA（通过神经基元混合实现可解释图学习）这一设计上即具可解释性的框架来解决这两个问题。在特征建模上，HARMONIA引入了神经基元混合（MoNB），将特征路由至特化的基元专家，从而在不牺牲特征特定特化的前提下实现参数共享。在结构建模上，HARMONIA使用相对随机游走概率（RRWP）来捕获多跳和多路径关系，并提出稀疏RRWP聚合（SRA），通过稀疏图传播计算这些交互，避免二次方成对复杂度。HARMONIA保留了简单的加性形式，其中预测分解为由结构影响调制的特征响应。实验上，HARMONIA在解释恢复方面强于现有可解释图基线，同时保持有竞争力的预测性能，并可扩展到拥有数百万节点的图。这些结果表明，可解释图学习在保持忠实性和可扩展性的同时，无需牺牲预测有效性。
英文摘要
Existing interpretable graph additive models still face limitations in either computational scalability or modeling flexibility. In terms of structural modeling, previous approaches either face quadratic scaling costs or sacrifice explicit source-to-target contribution decomposition. In terms of feature components, they rely either on per-feature neural networks or on single shared bases with limited feature specialization. We address both problems by introducing HARMONIA: Interpretable Graph Learning through Mixtures of Neural Bases, an interpretable-by-design framework. For feature modeling, HARMONIA introduces a Mixture of Neural Bases (MoNB), which routes features to specialized basis experts, enabling parameter sharing without sacrificing feature-specific specialization. For structural modeling, HARMONIA uses Relative Random Walk Probabilities (RRWP) to capture multi-hop and multi-path relationships, and proposes Sparse RRWP Aggregation (SRA) to compute these interactions through sparse graph propagation without quadratic pairwise complexity. HARMONIA retains a simple additive form in which predictions decompose into feature responses modulated by structural influence. Empirically, HARMONIA achieves stronger explanation recovery than existing interpretable graph baselines while maintaining competitive predictive performance and scaling to graphs with millions of nodes. These results show that interpretable graph learning can remain both faithful and scalable without sacrificing predictive effectiveness.
