部分因果效应识别的最优实验选择研究

View PDF HTML (experimental)

Abstract:Causal queries are often only partially identifiable from observational data, and experiments that could tighten the resulting bounds are typically costly. We study the problem of selecting, prior to observing experimental outcomes, a cost-constrained subset of experiments that maximally tightens bounds on a target query. We formalize this as the max-potency problem, where epistemic potency measures the worst-case reduction in bound width guaranteed by an experiment, and show that this problem is NP-hard via a reduction from 0-1 knapsack. Building on the polynomial-programming framework of Duarte et al. (2023), we give a general procedure for evaluating epistemic potency in discrete settings. To control the super-exponential search space, we introduce two graphical pruning criteria that depend only on the causal graph and the query: a novel path-interception rule that exploits district structure to certify zero potency in linear time, and an identifiability check based on the ID algorithm. On Erdos-Renyi random graphs and 11 bnlearn benchmark networks, the two criteria together prune 50-88% of candidate experiments on average without solving a single polynomial program. For the general subset search, we show that ID-pruned experiments are combinatorially inert, yielding a super-exponential reduction in the number of subsets evaluated. We close with an end-to-end demonstration on observational NHANES data, selecting optimal experiments for estimating the effect of physical activity on diabetes.

Comments:	Accepted at NeurIPS 2026
Subjects:	Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
Cite as:	arXiv:2605.06993 [cs.AI]
	(or arXiv:2605.06993v3 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2605.06993 arXiv-issued DOI via DataCite

Submission history

From: Tobias Maringgele [view email]
[v1] Thu, 7 May 2026 22:16:58 UTC (1,504 KB)
[v2] Sat, 5 Sep 2026 09:31:07 UTC (1,518 KB)
[v3] Tue, 27 Oct 2026 13:52:58 UTC (1,516 KB)