Composite Trust in Networked Multi-Agent Systems
John S. Baras, and Tao Jiang
Proceedings of the 2012 American Control Conference (ACC 2012), pp. 3547-3552, Montreal, Canada, June 27-29. 2012.
Trust plays a crucial role in the analysis, synthesis and
operation of the integrated system consisting of social-cognitive,
information and communication networks. Multi-agent systems
and associated coalition operations, which involve two or more
organizations, further augment the diversity and complexity of
network interactions. Trust and its derivative notions affect
dramatically the networked coalition operations. A substantial
part of the research challenge has to do with the multitude of
meanings, interpretations, symbolisms and mathematical
models used to represent and analyze trust. In this paper, we
introduce value directed graphs with weighted nodes as our
model for composite trust. We extend the value directed graph with weighted nodes composite trust model to include not only numerical weights, but also constraints. We show that the semiring-based constraint satisfaction problem (SCSPs) framework can serve as the unified model to investigate trust relation establishment.