提出新方法区分实体与关系的不确定性，显著提升知识图谱外分布检测能力
Decomposing Uncertainty in Probabilistic Knowledge Graph Embeddings: Why Entity Variance Is Not Enough
- 将不确定性分解为语义与结构两类，分别捕捉新实体和新关系
- 在时间分布偏移上实现0.94-0.99的AUROC，比原有方法提升60%-80%
- 适合需要高可信度推理的知识图谱应用，如医疗、金融等关键领域
概率知识图谱嵌入将实体表示为分布，用学习到的方差量化认知不确定性。我们发现其根本局限：方差与关系无关，导致同一实体在不同关系上下文中获得相同不确定性。这混淆了两种行为相反的分布外现象：新兴实体（罕见、学习不足）与新关系上下文（熟悉实体出现在未观测关系中）。我们证明一个不可能性结果：仅依赖与关系无关的实体级统计量的不确定性估计器，在新关系上下文中近乎随机地进行分布外检测。我们在三个数据集上验证，100%的新关系三元组存在频率匹配的分布内对应项。这解释了为何现有方法在随机篡改上达到0.99 AUROC，但在时间分布偏移上仅0.52-0.64。我们形式化地将不确定性分解为互补成分：来自实体嵌入方差的语义不确定性（检测新兴实体）和来自实体-关系共现的结构性不确定性（检测新关系）。主要理论结果证明两者非冗余，且任意凸组合严格优于单一信号。我们的方法（CAGP）通过学习权重融合两类不确定性，在多个基准上对时间分布偏移实现0.94-0.99的AUROC，相对基线提升60%-80%。实证验证确认三数据集（FB15k-237, WN18RR, YAGO3-10）完全频率重叠。在选择性预测中，该方法在85%答案率下将错误减少43%。
原文摘要 · Abstract (English)
Probabilistic knowledge graph embeddings represent entities as distributions, using learned variances to quantify epistemic uncertainty. We identify a fundamental limitation: these variances are relation-agnostic, meaning an entity receives identical uncertainty regardless of relational context. This conflates two distinct out-of-distribution phenomena that behave oppositely: emerging entities (rare, poorly-learned) and novel relational contexts (familiar entities in unobserved relationships). We prove an impossibility result: any uncertainty estimator using only entity-level statistics independent of relation context achieves near-random OOD detection on novel contexts. We empirically validate this on three datasets, finding 100 percent of novel-context triples have frequency-matched in-distribution counterparts. This explains why existing probabilistic methods achieve 0.99 AUROC on random corruptions but only 0.52-0.64 on temporal distribution shift. We formalize uncertainty decomposition into complementary components: semantic uncertainty from entity embedding variance (detecting emerging entities) and structural uncertainty from entity-relation co-occurrence (detecting novel contexts). Our main theoretical result proves these signals are non-redundant, and that any convex combination strictly dominates either signal alone. Our method (CAGP) combines semantic and structural uncertainty via learned weights, achieving 0.94-0.99 AUROC on temporal OOD detection across multiple benchmarks, a 60-80 percent relative improvement over relation-agnostic baselines. Empirical validation confirms complete frequency overlap on three datasets (FB15k-237, WN18RR, YAGO3-10). On selective prediction, our method reduces errors by 43 percent at 85 percent answer rate.
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。