RI: Small: Theory and Application of Mechanism Design for Team Formation

RI：小：团队形成机制设计理论与应用

• 批准号: 1526860
• 负责人: Yevgeniy Vorobeychik
• 金额: $ 44.21万
• 依托单位: Vanderbilt University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-01 至 2019-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1526860&HistoricalAwards=false
• 关键词: RI; Small; Theory; Application; Mechanism

In a world where complexity of tasks abounds, teamwork is the norm rather than the exception. Formation of effective teams is therefore vital in nearly any organization, and across a broad array of domains, such as large-scale software projects, course projects, military operations, construction projects, and search-and-rescue tasks. Successful team formation, however, hinges on having knowledge about how individuals working on a team complement each other in service of a task, their ability to work together, as well as potentially misaligned preferences about goals that the different tasks may achieve. The goal of this project is to develop a general-purpose team formation exchange service, leveraging new theoretical foundations for designing team formation mechanisms when the abilities and preferences of prospective teammates are uncertain. Such an exchange would significantly improve the ability of organizations to dynamically form agile and efficient project teams, reducing conflict and better aligning individual and organizational incentives. This project considers an exceptionally challenging conceptual and technical problem of mechanism design in the context of team formation in service of a set of goals. The core conceptual challenge lies in modeling the problem in sufficient detail to capture considerations such as achievement of sub-goals by tasks, agent capabilities, task requirements, and agent preferences about goals and teammates, but avoiding unnecessary detail to maintain tractability. The proposed research offers a novel modeling framework aimed at balancing these two considerations. The technical contributions involve both novel theoretical treatment of the problem, such as characterizations of novel extensions of mechanisms from related domains, including combinatorial assignment, entirely new mechanisms tailored to the problem at hand, and algorithmic contributions to optimal scalable automated mechanism design leveraging specific problem structure. Additionally, the proposed research will blend experimental economics research with contributions in theoretical and computational economics by using human subject experiments as one of the means to evaluate the developed mechanisms.

在一个任务复杂的世界里，团队合作是常态而不是例外。因此，在几乎任何组织中，组建有效的团队都是至关重要的，并且跨越了广泛的领域，例如大型软件项目，课程项目，军事行动，建筑项目和搜索和救援任务。然而，成功的团队组建取决于了解团队中的个人如何在任务中相互补充，他们合作的能力，以及对不同任务可能实现的目标的潜在不一致偏好。本项目的目标是开发一个通用的团队形成交流服务，利用新的理论基础，设计团队形成机制时，未来的队友的能力和偏好是不确定的。 这种交流将大大提高各组织的能力，以动态地组建灵活和高效的项目团队，减少冲突，更好地协调个人和组织的激励措施。该项目认为，一个非常具有挑战性的概念和技术问题的机制设计的背景下，团队的形成服务的一套目标。核心的概念性挑战在于建模的问题，在足够的细节，以捕获的考虑因素，如子目标的实现任务，代理的能力，任务要求，代理的偏好目标和队友，但避免不必要的细节，以保持易处理性。拟议的研究提供了一个新的建模框架，旨在平衡这两个方面的考虑。技术贡献涉及两个新的理论处理的问题，如表征的新扩展的机制，从相关领域，包括组合分配，全新的机制，适合手头的问题，和算法的贡献，最佳的可扩展的自动化机制设计利用特定的问题结构。此外，拟议的研究将通过使用人类受试者实验作为评估所开发机制的手段之一，将实验经济学研究与理论和计算经济学的贡献相结合。

项目成果

Robust Linear Regression Against Training Data Poisoning

• DOI: 10.1145/3128572.3140447
• 发表时间: 2017-11
• 期刊: Proceedings of the 10th ACM Workshop on Artificial Intelligence and Security
• 作者: Chang Liu;Bo Li;Yevgeniy Vorobeychik;Alina Oprea

A Decision Theoretic Framework for Emergency Responder Dispatch

紧急救援人员调度的决策理论框架

• 发表时间: 2018
• 期刊: International Conference on Autonomous Agents and Multiagent Systems
• 作者: Mukhopadhyay, Ayan;Wang, Zilin;Vorobeychik, Yevgeniy
• 通讯作者: Vorobeychik, Yevgeniy

Near-optimal interdiction of factored MDPs

因子 MDP 的近乎最佳拦截

• 发表时间: 2017
• 期刊: Conference on Uncertainty in Artificial Intelligence
• 作者: Panda, Swetasudha;Vorobeychik, Yevgeniy
• 通讯作者: Vorobeychik, Yevgeniy

Controlling Elections through Social Influence

通过社会影响力控制选举

• 发表时间: 2018
• 期刊: International Conference on Autonomous Agents and Multiagent Systems
• 作者: Wilder, Bryan;Vorobeychik, Yevgeniy
• 通讯作者: Vorobeychik, Yevgeniy

Prioritized Allocation of Emergency Responders based on a Continuous-Time Incident Prediction Model

• 发表时间: 2017-05
• 作者: Ayan Mukhopadhyay;Yevgeniy Vorobeychik;A. Dubey;Gautam Biswas