DDRIG in DRMS: Multi-target Technology Deployment and Information Disclosure in Attacker-defender Settings: Analyzing Game-theoretic Prescriptions and Human Decisions

DRMS 中的 DDRIG：攻击者-防御者设置中的多目标技术部署和信息披露：分析博弈论处方和人类决策

• 批准号: 2215097
• 负责人: Jun Zhuang
• 金额: $ 1.53万
• 依托单位: SUNY at Buffalo
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-08-01 至 2023-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2215097&HistoricalAwards=false
• 关键词: DDRIG; DRMS; Multi; target; Technology

Defending against adversarial threats has been a central and longstanding focus for governments throughout the world. One prominent approach to improve defensive capabilities is to deploy new technologies among venues of interest (e.g., metal detectors at airports). When it comes to deploying new security and defense technologies, agencies must decide how the related information should be released to the public. Given that adversaries, such as terrorist organizations, can access information that is publicly disseminated, it is critical to understand the implications of releasing different types of information. For instance, security agencies may choose to release information on only a subset of venues where new technology is deployed, creating an uncertainty for an adversary regarding the deployment of the technology at other venues (e.g., airports). This research addresses this difficult information disclosure problem by building on previous work in game theory and adversarial decision making to develop models and experiments that provide insights into defensive information disclosure, adversarial beliefs, and adversarial target selection decisions.The research objectives of this proposed effort are to mathematically model, experimentally test, and robustly analyze technology deployment and information disclosure strategies in the context of security and defense. To meet these objectives, this project develops and analyzes a novel game-theoretic signaling model, which provides insightful analyses into optimal information disclosure strategies. Further, human experiments –designed to mimic the game model – are conducted to (i) study human beliefs and decision making in this context, and (ii) compare the game-theoretic prescriptions with actual human decisions. Successful completion of this project helps to inform the long-term effectiveness of technology deployments within the security and defense sectors. Although this research is motivated by problems in security and defense, the mathematical model and experimental framework is generalizable to any application in which the strategic release of information across multiple venues is of interest.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.

抵御敌对威胁一直是世界各国政府长期关注的中心问题。提高防御能力的一个重要方法是在相关场所（例如机场的金属探测器）部署新技术。当涉及到部署新的安全和防御技术时，各机构必须决定如何向公众发布相关信息。鉴于对手，如恐怖组织，可以访问公开传播的信息，理解发布不同类型信息的含义是至关重要的。例如，安全机构可能选择只发布部署了新技术的一小部分场所的信息，这给对手在其他场所（例如机场）部署该技术带来了不确定性。本研究通过建立在博弈论和对抗性决策方面的先前工作的基础上，开发模型和实验来解决这一困难的信息披露问题，为防御性信息披露、对抗性信念和对抗性目标选择决策提供见解。本文的研究目标是建立安全与防御背景下的技术部署和信息披露策略的数学模型、实验测试和稳健分析。为了实现这些目标，本项目开发并分析了一个新的博弈论信号模型，该模型为最优信息披露策略提供了深刻的分析。此外，人类实验——旨在模仿游戏模型——被用于(i)研究人类在这种情况下的信念和决策，以及（ii）将博弈论处方与实际的人类决策进行比较。该项目的成功完成有助于为安全和国防部门技术部署的长期有效性提供信息。虽然这项研究的动机是在安全和防御问题，数学模型和实验框架是推广到任何应用中，其中跨多个场所的战略信息发布感兴趣。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。