论文概要
研究领域: ML 作者: Bijan Mazaheri, Jiaqi Zhang, Caroline Uhler 发布时间: 2026-07-13 arXiv: 2607.11816
中文摘要
因果发现算法学习描述随机变量间因果依赖的网络。常见工作流程首先利用观测数据上的条件独立性性质确定部分定向的因果关系，然后应用干预来定向未知因果方向。第一步的关键假设是忠实性：要求因果关联变量表现出统计依赖性。许多自然系统包含缓冲和稳定路径，相互抵消以实现系统鲁棒性。这种路径抵消违反忠实性，导致因果发现算法错误地移除因果依赖。本文认为硬干预包含关于因果联动存在/缺失的信息，在结构发现的第一阶段被忽视了。证明一个温和假设——干预即时忠实性——允许抵消，足以用硬干预非参数化识别因果结构。这些结果将干预定位为因果结构信息的主要载体，应优先于条件独立性检验。为翻转范式，还指定了当识别标准因干预范围限制而未满足时的等价类。
原文摘要
Causal discovery algorithms learn a network that describes the causal dependencies among random variables. A common workflow involves first utilizing conditional independence properties on observational data to determine partially directed causal relationships, then applying interventions to orient the unknown causal directions. A critical assumption for the first step is faithfulness: a requirement that causally linked variables exhibit statistical dependence. Many natural systems include buffering and stabilizing pathways that cancel out to achieve systemic robustness. This cancellation of pathways violates faithfulness, leading causal discovery algorithms to incorrectly remove causal dependencies. In this paper, we argue that hard interventions contain information about the presence/abs…
— 自动采集于 2026-07-15
#论文 #arXiv #ML #小凯