III: Small: pCAR: Discovering and Leveraging Plausibly Causal (p-causal) Relationships to Understand Complex Dynamic Systems

今天，要成功应对社会经济关键领域(如可持续性、公共卫生和生物学)的许多紧迫挑战，需要从高维数据中更深入地了解各种实体之间的因果关系和相互作用。例如，流行病的出现和传播涉及各种实体之间的因果复杂相互作用，包括个人及其社会互动、实际的短期和长期流动网络，以及决策者的干预决定(如关闭学校或限制流动)。为了填补这一重要漏洞，并使具有重大经济和公共卫生影响的应用程序和服务成为可能，该项目开展了具有因果意识的数据驱动科学和工程研究，包括数据支持的因果分析。该项目还为劳动力增长做出了贡献，为博士、硕士和本科水平的研究和教学提供了良好的环境。同时拥有数据分析和管理技能的未来研究人员正在接受培训，学生们将通过参与研究、发表论文和与领域专家合作来了解职业道路。在这个项目中，研究团队假设数据分析提供了机会来识别可能是因果关系。他们进一步假设，数据可以用来加强和削弱因果假设，也可以用来修剪那些不存在因果关系的关系。为了验证和利用这些假说，他们引入了“合理因果关系”(p-因果关系)和“合理因果关系(p-因果关系)”的新概念，他们开发了一些技术来(I)发现p-因果关系的相互作用和关系，(Ii)在因果关系本身随时间演变的系统中保持这些p-因果关系，以及(Iii)使用发现的p-因果关系来支持高效和有效的数据分析，以研究复杂的动态系统。特别是，他们开发了新的模型来捕捉复杂、动态系统中与上下文相关的看似合理的因果(p-因果)关系，并设计了新颖且可扩展的因果数据分析算法，该算法利用不同和潜在演变的上下文中实体之间的已知或假设的p-因果关系来处理数据稀疏性、不精确性和噪声。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Today, successfully tackling many urgent challenges insocio-economically critical domains (such as sustainability, publichealth, and biology) requires obtaining a deeper understanding of causalrelationships and interactions among a diverse spectrum of entities fromhigh-dimensional data. For example, the emergence and propagation of anepidemic involves a causally complex interplay of entities, includingindividuals and their social-interactions, physical short-range andlong-range networks of mobility, and intervention decisions (such asschool closures or restrictions on mobility) by decision makers. Withthe aim of filling this important hole and enabling applications andservices with significant economic and public health impact, the project carriesout research on causally-aware data-driven science and engineering,including data-supported causal analysis. The project also contributesto workforce growth, provides an excellent context for doctoral,master's, and undergraduate level research and teaching. Futureresearchers that have the skills in both data analysis and managementare being trained and students are being introduced to career pathwaysthrough their participation in research, publications, and partnershipswith domain experts.In this project, the research team hypothesizes that data analysis providesopportunities for identifying relationships that can potentially becausal. They further hypothesize that data can be used for strengtheningand weakening causal assumptions, and for pruning relationships that arecertainly not causal. To validate and leverage these hypotheses, theyintroduce the novel concepts of 'plausible causality' (p-causality) and'plausibly causal (p-causal) relationships' and they developtechniques to (i) discover p-causal interactions and relationships, (ii)maintain these p-casual relationships in systems where causality itselfevolves over time, and (iii) use discovered p-causal relationships tosupport efficient and effective data analytics to study complex, dynamicsystems. In particular, they develop new models to capturecontext-sensitive plausibly causal (p-causal) relationships in complex,dynamic systems and design novel and scalable causally-awaredata analysis algorithms that leverage known or hypothesized p-causalrelationships among entities within different and potentially evolvingcontexts to deal with data sparsity, imprecision, and noise.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.