arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.08508cs.LG
收藏

缓解有符号图神经网络中的度偏差

Mitigating Degree Bias in Signed Graph Neural Networks

Huazhong Agricultural University(华中农业大学)

Fang He, Jinhai Deng, Ruizhan Xue, Maojun Wang, Zeyu Zhang

AI总结:

针对有符号图神经网络存在的度偏差公平性问题,提出模型无关的DD-SGNN方法,通过头尾三元组内高低度节点的信息转移补充尾节点结构并保留正负语义,在四个真实数据集上实现去偏且不降低AUC、F1性能。

摘要:

与图神经网络(GNNs)一样,有符号图神经网络(SGNNs)也面临着来自源数据和典型聚合方法的公平性问题。本文率先将GNNs中的公平性研究拓展至SGNNs领域。我们识别出有符号图中的度偏差问题,为SGNNs相关的公平性问题提供了新的研究视角。为解决这一偏差问题,受以往度偏差相关研究的启发,本文提出了一种新的模型无关方法,用于增强不同度数节点的表示,该方法被命名为Degree Debiased Signed Graph Neural Network(DD-SGNN,度去偏有符号图神经网络)。更具体地说,在每一层中,我们在头尾三元组内实现从高度数节点到低度数节点的信息转移,以此补充尾节点缺失的潜在域结构,同时保留有符号图中平衡理论所规定的正负语义。我们在四个真实世界数据集上开展了大量实验。结果验证了该模型的有效性,即我们的模型在不牺牲性能(即AUC、F1)的前提下缓解了度偏差问题。代码已在补充材料中提供。

英文摘要:

Like Graph Neural Networks (GNNs), Signed Graph Neural Networks (SGNNs) are also up against fairness issues from source data and typical aggregation method. In this paper, we are pioneering to make the investigation of fairness in SGNNs expanded from GNNs. We identify the issue of degree bias within signed graphs, offering a new perspective on the fairness issues related to SGNNs. To handle the confronted bias issue, inspired by previous work on degree bias, a new Model-Agnostic method is consequently proposed to enhance representation of nodes with different degrees, which named as Degree Debiased Signed Graph Neural Network (DD-SGNN) . More specifically, in each layer, we make a transfer from nodes with high degree to nodes with low degree inside a head-to-tail triplet, which to supplement the underlying domain missing structure of the tail nodes and meanwhile maintain the positive and negative semantics specified by balance theory in signed graphs. We make extensive experiments on four real-world datasets. The result verifies the validity of the model, that is, our model mitigates the degree bias issue without compromising performance($\textit{i.e.}$, AUC, F1). The code is provided in supplementary material.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑