Dynamic Evolution of Complex Networks: A Reinforcement Learning Approach Applying Evolutionary Games to Community...

Area of Science:

• Complex Systems Science
• Network Science
• Computational Social Science

Background:

• Current studies on dynamic systems lack models for individual birth-death and community development.
• Understanding emergent structures in complex systems requires integrating individual behaviors with network dynamics.

Purpose of the Study:

• To propose a novel networked evolution model incorporating birth-death processes and reinforcement learning.
• To investigate the emergence and evolution of cooperative behaviors and community structures.
• To validate the model's practicality using real-world data.

Main Methods:

• Developed a networked evolution model with individual birth-death, Q-learning reinforcement learning, and spatial movement.
• Simulated systems with and without birth-death processes to observe behavioral and structural evolution.
• Validated model fitting with real-world population and network data.

Main Results:

• The model successfully reproduces cooperative behaviors and community structures.
• Exploitation rates and payoff parameters were identified as key drivers for community emergence.
• Learning rates, discount factors, and spatial dimensions influence community formation speed, stability, and size.

Conclusions:

• The proposed model offers a new perspective on community development in dynamic systems.
• It provides a robust framework for studying population dynamics and emergent network structures.
• The model's parameters offer insights into factors governing community formation and stability.