CAREER: A Holistic Framework for the Analysis of Information Dynamics in Human Networks

点击翻译按钮获取中文摘要

Communications between humans is different from communications between machines due to cognitive limitations, biases, and misaligned objectives, yet human behavior is still predictable in a statistical sense. Motivated by this observation, this project investigates problems of theoretical and practical significance concerning human communication networks. Typical scenarios include human-machine systems and social networks where selfish agents, some of whom are prone to bias and limitations, control nodes to manage information flow to maximize their own perceived payoff. While relevant models have been studied in the field of Behavioral Economics, their extensions to realistic communication and control settings pose many open problems. This project will adopt a holistic approach that considers an end-to-end problem jointly from information, control, and game-theoretic perspectives, leveraging tools and models from all three disciplines. A notable application of the developed framework is on modeling public opinion polarization and misinformation spread over social networks. Another application pertains to mitigating the impact of bias in information processing systems. The work will address fundamental questions regarding the emerging human communication paradigm. This growing research field requires revisiting key results in classical communication and control theories and poses significant challenges, requiring approaches and tools from multiple disciplines. The research outcomes are expected to constitute an essential step in understanding the interplay of game theory and economics with information theory, control, communications, and compression. Specific goals of this project are categorized into three groups. The first set of goals investigates communication and compression problems between agents with misaligned objectives, cognitive limitations, and biases. The second set of goals concerns behavioral linear averaging models of opinion evolution and their applications in polarization dynamics. The final set of goals explores truthful data gathering strategies in behaviorally biased information networks.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.