Using synchronous Boolean networks to model several phenomena of collective behavior

Area of Science:

• Complex Systems
• Computational Social Science
• Network Science

Background:

• Collective behavior in multi-agent systems involves agents transitioning between states.
• Agent behavior can be conforming, where current actions align with past agent actions.
• Synchronous Boolean networks provide a framework for modeling these discrete state transitions.

Purpose of the Study:

• To develop an approach for modeling and analyzing collective behavior in multi-agent systems.
• To investigate the roles of specific agent types (instigators, loyalists) in state transitions.
• To address combinatorial problems related to controlling collective states.

Main Methods:

• Modeling collective behavior using synchronous Boolean networks.
• Defining agent types: instigators (always active), loyalists (never active), and simple agents.
• Formulating combinatorial problems for controlling network states via instigators and loyalists.
• Reducing problems to the Boolean satisfiability (SAT) problem for computational solutions.

Main Results:

• Theoretical results on the behavior of agents with conforming and anticonforming strategies.
• Successful computational experiments on randomly generated networks with hundreds of vertices.
• Demonstrated the efficacy of SAT solvers in solving the defined combinatorial problems.

Conclusions:

• The proposed approach effectively models and analyzes collective behavior in multi-agent systems.
• Specific agent configurations (instigators, loyalists) can deterministically control system states.
• The reduction to SAT provides a scalable method for solving complex control problems in these networks.