arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1207.1389cs.AIstat.ME
收藏

关于确定N个变量间所有因果关系的充分和最坏情况下所需的试验次数

On the Number of Experiments Sufficient and in the Worst Case Necessary to Identify All Causal Relations Among N Variables

Carnegie Mellon University(卡内基梅隆大学)

Frederick Eberhardt, Clark Glymour, Richard Scheines

摘要:

我们表明,如果任何数量的变量被允许在任何一个实验中同时和独立随机化,log 2(N)+1实验是足够的,在最坏的情况下,必要的,以确定N >= 2个变量之间的因果关系时,没有潜变量,没有样本选择偏差,没有反馈周期。对于所有K,0 < K < 1/(2N),我们提供了当每个实验同时随机化K个变量时确定因果结构所需的实验数量的上限。对于较大的N,这些界限显着低于N - 1界限时,每个实验随机化最多一个变量所需的。当kmax < N/2时,我们证明了(N/kmax-1)+N/(2kmax)log 2(kmax)实验是充分的,在最坏情况下是必要的。我们在猜想的最小数量的实验,在最坏的情况下,足以确定所有的因果关系N个观察到的变量是一个DAG的顶点的子集。

英文摘要:

We show that if any number of variables are allowed to be simultaneously and independently randomized in any one experiment, log2(N) + 1 experiments are sufficient and in the worst case necessary to determine the causal relations among N >= 2 variables when no latent variables, no sample selection bias and no feedback cycles are present. For all K, 0 < K < 1/(2N) we provide an upper bound on the number experiments required to determine causal structure when each experiment simultaneously randomizes K variables. For large N, these bounds are significantly lower than the N - 1 bound required when each experiment randomizes at most one variable. For kmax < N/2, we show that (N/kmax-1)+N/(2kmax)log2(kmax) experiments aresufficient and in the worst case necessary. We over a conjecture as to the minimal number of experiments that are in the worst case sufficient to identify all causal relations among N observed variables that are a subset of the vertices of a DAG.

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。

↑