Spatio-Temporal Graphical Counterfactuals: An Overview
School of Mathematics, Southeast University, Nanjing 210096 , China(数学学院,东南大学,南京);School of Transportation, Southeast University, Nanjing 210096 , China(交通运输学院,东南大学,南京);Department of Computer Science, Rensselaer Polytechnic Institute, Troy NY 12180 , USA(计算机科学系,罗切斯特理工学院,特洛伊);Frontiers Science Center for Mobile Information Communication(移动信息通信前沿科学中心);Purple Mountain Laboratories, Nanjing 211102 , China(紫金山实验室,南京)
Counterfactual thinking is a crucial yet challenging topic for artificial intelligence to learn knowledge from data and ultimately improve performance for new scenarios. Many research works, including the Potential Outcome Model (POM) and the Structural Causal Model (SCM), have been proposed to address this. However, their modeling, theoretical foundations, and application approaches often differ. Moreover, there is a lack of graphical approaches for inferring spatio-temporal counterfactuals, that account for spatial and temporal interactions among multiple units. Thus, in this work, we aim to present a survey that compares and discusses different counterfactual models, theories and approaches. Additionally, we propose a unified graphical causal framework to infer spatio-temporal counterfactuals.
Counterfactual Analysis of the Impact of the IMF Program on Child Poverty in the Global-South Region using Causal-Graphical Normalizing Flows
Linköping University(林雪平大学)
这项工作演示了因果推理和深度学习模型的一个特定分支的应用:\emph{causal-Graphical Normalizing Flows (c-GNFs)}。在最近的一篇文章中,学者们表明,规范化流具有某些属性,使其特别适合于因果分析和反事实分析。然而,c-GNF仅在模拟数据环境中进行了测试,迄今为止还没有对c-GNF在大规模真实世界数据中的应用进行过评估。以\emph{AI for social good}为重点,我们的研究提供了一个反事实分析的影响,国际货币基金组织(IMF)计划对儿童贫困使用c-GNF。该分析依赖于大规模的真实世界观察数据:1,941,734名18岁以下的儿童,由居住在全球南部67个国家的567,344个家庭照顾。虽然国际货币基金组织的主要目标是支持各国政府实现经济稳定,但我们的研究结果发现,国际货币基金组织的一项计划将儿童贫困作为一种积极的副作用减少了约1.2 $\pm$ 0.24度(“0”等于没有贫困,“7”是最大贫困)。因此,我们的文章展示了c-GNF如何进一步在AI中使用深度学习和因果推理来实现社会公益。它展示了如何使用学习算法通过在群体水平(ACE)、亚群体水平(CACE)和个体水平(ICE)的反事实推理来解决未开发的潜在重大社会影响。与大多数模型ACE或CACE而不是ICE的作品相反,c-GNF使用\emph{`The First Law of …
Strategic Counterfactual Modeling of Deep-Target Airstrike Systems via Intervention-Aware Spatio-Causal Graph Networks
Dhurakij Pundit University(泰国易三仓大学);The University of Western Australia(西澳大利亚大学);Association for Computing Machinery(美国计算机协会);Royal Anthropological Institute(皇家人类学学会)
A General Neural Causal Model for Interactive Recommendation
Computer Network Information Center, Chinese Academy of Sciences(中国科学院计算机网络信息中心);University of Chinese Academy of Sciences(中国科学院大学);Chinese University of Hong Kong(香港中文大学);City University of Hong Kong(香港城市大学)