弱忠实性的操作化:恢复、歧义与计算极限
Operationalising Weaker Faithfulness: Recovery, Ambiguity and Computational Limits
Cillian Hourican · Rick Quax
中文摘要
变量可能共同预测结果,却各自与结果独立,使结构学习器丢弃有用联系。研究通过两种方式将弱忠实性操作化:在邻接学习器中选择性保护,以及显式保留替代有向图解释。对照实验表明,早期候选移除和后期剪枝两处改动可能都需要。在带后代节点的噪声 XOR(N = 10,000,20 个匹配样本)中,结合重新激活与保护后,完整父节点对恢复率从任一单独改动的 0/20 提升至 20/20,但会加入错误边。确认研究表明,家族估计器能保留因定向受保护学习器所选骨架而丢失的解释:在较弱信号 XOR 条件下(N = 10,000),50 个样本中有 48 个保留完整参考家族,而骨架适配器为 0/50。这些收益取决于条件;保护可能只增加错误,较弱效果会造成错误排除,普通忠实链比较则有利于 PC 学习器。变量数为 6/8 时,完整非空家族输出减少;截至报告截止时间,全部 400 次 12/20 变量搜索仍未完成。神经对照显示工程化保留机制,而家族估计器在应用预算内未返回可接受图。该操作化明确了示例中必须协调的两个删除阶段,并展示了直接家族搜索何时能保留选择单一骨架所丢失的解释。有限证据与计算仍是可靠泛化推断的重要障碍。
关键要点
- 01弱忠实性下,变量可共同预测结果,但单独看似独立,导致有用联系被丢弃
- 02结合重新激活与选择性保护,可将完整父节点对恢复率由 0/20 提升至 20/20
- 03家族估计器在 48/50 个样本中保留完整参考家族,优于骨架适配器的 0/50
- 04保护可能引入错误边,弱信号会导致错误排除,PC 学习器在普通比较中更优
- 05高变量数搜索未完成,家族估计器也未能在预算内返回可接受图
解读
尚无解读。
原始英文摘要
arXiv:2610.31724v1 Announce Type: cross Abstract: Variables can predict an outcome jointly while each is individually independent of it, causing structure learners to discard informative links. We investigate how to operationalise weaker faithfulness through two approaches: selective protection in an adjacency learner and explicit retention of alternative directed graph explanations. Matched experiments show that changes at both an early candidate-removal stage and later pruning can be necessary. In noisy XOR with a descendant (N = 10,000, 20 matched samples), combining reactivation and protection raises complete parent-pair recovery from 0/20 for either change alone to 20/20, while adding false edges. A confirmation study shows that the family estimator can preserve explanations lost by orienting the protected learner's selected skeleton: in one weaker signal XOR condition (N = 10,000) it retains the entire reference family in 48/50 samples, compared with 0/50 for the skeleton adapter. These gains are conditional. Protection can add only error, weaker effects produce false exclusions, and an ordinary faithful-chain comparison favours the PC learner. Complete nonempty family outputs decline at six/eight variables, and all 400 twelve/twenty-variable searches remain unfinished at the reporting cutoff. Neural controls show engineered retention, while the family estimator returns no accepted graph within its application budget. This operationalisation identifies two deletion stages that must be coordinated in the motivating example and shows when direct family search preserves explanations lost by selecting one skeleton. Finite evidence and computation remain substantial barriers to reliable general inference.