📝 SummarXiv

Abstract

Collaboration is a fundamental and essential characteristic of many complex systems, ranging from ant colonies to human societies. Each component within a complex system interacts with others, even at a distance, to accomplish a given task. A network of collaboration can be defined to study the collective behavior of such systems within the framework of complex networks. The nodes in these networks may represent simple organisms or more sophisticated intelligent agents, such as humans. In this study, we utilize intelligent agents (nodes) trained through reinforcement learning techniques to establish connections with their neighbors, ultimately leading to the emergence of a large-scale communication cluster. Notably, there is no centralized administrator; instead, agents must adjust their connections based on information obtained from local observations. The connection strategy is formulated using a physical Hamiltonian, thereby categorizing this intelligent system under the paradigm of "Physics-Guided Machine Learning". The resulting self-organized distributed complex network has numerous industrial applications, including constructing Internet of Things (IoT) networks. The design of such networks often encounters challenges, the most critical of which is ensuring effective connectivity for reliable communication while optimizing energy consumption. IoT networks are inherently dynamic in many real-world applications, such as Vehicle Ad-hoc Networks (VANETs), where nodes are mobile, and the connection topology evolves rapidly over time. These systems require a robust and rapidly self-organizing communication network. Our findings demonstrate that the proposed intelligent agents facilitate the formation of self-organized complex networks capable of maintaining network-wide connectivity across various dynamic scenarios while simultaneously optimizing average electrical power consumption.