Dynamical quenching and annealing in self-organization multiagent models

Area of Science:

• Complex systems
• Agent-based modeling
• Statistical physics

Background:

• The generalized minority game (GMG) and bar attendance model (BAM) are agent-based models used to study self-organization and decision-making.
• Understanding the dynamics and asymptotic states of these models is crucial for various fields, including economics and social sciences.

Purpose of the Study:

• To compare the dynamics of the generalized minority game (GMG) with a modified bar attendance model (BAM).
• To investigate the asymptotic states (thermodynamic equilibrium and quenched configurations) in both models.
• To introduce an annealing protocol to transition GMG from quenched states to equilibrium states.

Main Methods:

• Agent-based simulation of GMG and BAM dynamics.
• Comparative analysis of model behaviors under different parameter settings, particularly the prize-to-fine ratio (G).
• Development and application of a simulated annealing protocol for GMG.

Main Results:

• Both GMG and BAM can exhibit thermodynamic equilibrium and quenched configurations, but under different conditions.
• The parameter G significantly influences the emergent states in the GMG.
• The proposed annealing protocol successfully guides GMG towards an equilibrium state comparable to that of the BAM.

Conclusions:

• The BAM offers a more information-efficient dynamic compared to the standard GMG dynamics.
• The study demonstrates a method to bridge the gap between different dynamic regimes within agent-based models.
• Findings provide insights into controlling and understanding emergent behavior in self-organizing systems.