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 2402.11494cs.LGcs.SI
收藏

基于因果干预的图分布外泛化

Graph Out-of-Distribution Generalization via Causal Intervention

Shanghai Jiao Tong University(上海交通大学)

Qitian Wu, Fan Nie, Chenxiao Yang, Tianyi Bao, Junchi Yan

AI总结:

针对图神经网络在分布偏移下性能下降且缺环境标签的问题,提出基于因果推断、协调环境估计器与混合专家GNN的方法,抵消混淆偏差提升泛化,在基准上准确率最高提升27.4%。

摘要:

分布外(OOD)泛化在图学习领域受到越来越多的关注,因为图神经网络(GNNs)在分布偏移情况下往往会出现性能下降。该问题的挑战在于,图上的分布偏移涉及节点间复杂的相互连接,且数据中通常缺少环境标签。本文采用自底向上的数据生成视角,通过因果分析揭示了一项关键发现:GNN在OOD泛化中失效的核心在于来自环境的潜在混淆偏差。这种偏差会误导模型利用自我图特征与目标节点标签之间的环境敏感相关性,导致模型在未见的新节点上泛化效果不佳。基于这一分析,我们提出了一种概念简单但具备理论依据的方法,用于在节点级分布偏移下训练鲁棒的GNN,且无需环境标签的先验知识。我们的方法采用了一种源自因果推断的新型学习目标,该目标协调了环境估计器与混合专家GNN预测器。这种新方法能够抵消训练数据中的混淆偏差,助力学习可泛化的预测关系。大量实验表明,我们的模型能够在多种类型的分布偏移下有效提升泛化能力,在图OOD泛化基准上相比现有最优方法准确率最高提升27.4%。源代码可在https://github.com/fannie1208/CaNet获取。

英文摘要:

Out-of-distribution (OOD) generalization has gained increasing attentions for learning on graphs, as graph neural networks (GNNs) often exhibit performance degradation with distribution shifts. The challenge is that distribution shifts on graphs involve intricate interconnections between nodes, and the environment labels are often absent in data. In this paper, we adopt a bottom-up data-generative perspective and reveal a key observation through causal analysis: the crux of GNNs' failure in OOD generalization lies in the latent confounding bias from the environment. The latter misguides the model to leverage environment-sensitive correlations between ego-graph features and target nodes' labels, resulting in undesirable generalization on new unseen nodes. Built upon this analysis, we introduce a conceptually simple yet principled approach for training robust GNNs under node-level distribution shifts, without prior knowledge of environment labels. Our method resorts to a new learning objective derived from causal inference that coordinates an environment estimator and a mixture-of-expert GNN predictor. The new approach can counteract the confounding bias in training data and facilitate learning generalizable predictive relations. Extensive experiment demonstrates that our model can effectively enhance generalization with various types of distribution shifts and yield up to 27.4\% accuracy improvement over state-of-the-arts on graph OOD generalization benchmarks. Source codes are available at https://github.com/fannie1208/CaNet.

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。

↑