基于准马尔可夫模型的因果概率与根本原因分析
机器学习
2025-09-03 v1 机器学习
摘要
因果概率提供了原则化的方法来评估因果关系,但由于部分可识别性和潜在混杂问题,面临计算挑战。本文引入了算法简化,显著降低了计算 tighter bounds(更紧密界限)的复杂度,并提出了一种新方法论框架,用于根本原因分析,通过系统性地运用这些因果指标对整个因果路径进行排序。
引用
@article{arxiv.2509.02535,
title = {Probabilities of Causation and Root Cause Analysis with Quasi-Markovian Models},
author = {Eduardo Rocha Laurentino and Fabio Gagliardi Cozman and Denis Deratani Maua and Daniel Angelo Esteves Lawand and Davi Goncalves Bezerra Coelho and Lucas Martins Marques},
journal= {arXiv preprint arXiv:2509.02535},
year = {2025}
}
备注
Accepted at the 35th Brazilian Conference on Intelligent Systems (BRACIS 2025)