Competition and Collaboration in Complex Networks
Abstract
The behavior of interacting agents across a wide range of domains is increasingly mediated by complex networks that link these agents together and create connections among their behaviors. This principle applies at large scales for example, in the ways in which large populations react to news, political messages, and the introduction of new technologies and at intermediate scales, when individuals in an organization must find ways of coordinating and collaborating in the face of potentially conflicting incentives. The field s understanding of the ways in which the structure of the underlying network affects these phenomena is limited by a lack of general techniques that can relate network properties to models of agent level behavior. The proposed research objective is to bridge this gap, through new approaches that combine algorithms and networks with game theoretic models of interactions. This overall objective is composed of several specic technical goals, based on domains in which the link between network structure and game theoretic interaction is particularly salient. One main goal is to develop game theoretic models for the flow of information through networks. A line of research has studied how information spreads in networks, but it has been primarily focused on the properties of individual pieces of information in isolation. The PIs will investigate the setting in which multiple pieces of potentially conflicting information spread in a network, controlled by entities with potentially opposing priorities. A second goal is to expand the power of current techniques for deriving game theoretic outcomes from dynamic behavior, especially focusing on the spontaneous emergence of cooperation in repeated strategic interactions. Many of the canonical predictions of game theory provide very little guidance to assess how an outcome might be reached through the properties of agent interaction over time.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 14, 2022
- Source ID
- FA95501910183
Entities
People
- Éva Tardos
Organizations
- Air Force Office of Scientific Research
- Cornell University
- United States Air Force