Modeling Complex Networks Based on Deep Reinforcement Learning

Abstract

The network topology of complex networks evolves dynamically with time. How to model the internal mechanism driving the dynamic change of network structure is the key problem in the field of complex networks. The models represented by WS, NW, BA usually assume that the evolution of network structure is driven by nodes’ passive behaviors based on some restrictive rules. However, in fact, network nodes are intelligent individuals, which actively update their relations based on experience and environment. To overcome this limitation, we attempt to construct a network model based on deep reinforcement learning, named as NMDRL. In the new model, each node in complex networks is regarded as an intelligent agent, which reacts with the agents around it for refreshing its relationships at every moment. Extensive experiments show that our model not only can generate networks owing the properties of scale-free and small-world, but also reveal how community structures emerge and evolve. The proposed NMDRL model is helpful to study propagation, game, and cooperation behaviors in networks.

Exploring the Polycentric Structure and Driving Mechanism of Urban Regions From the Perspective of Innovation Network

“…Such as Beijing is not only a knowledge production center and innovation linkage hub of BTH, but also a national or global knowledge production center or internal and external innovation linkage hub and gateway. Therefore, future research could further focus on the influence of the multiplicity of city networks at different scales on the formation and evolution of polycentric structures to provide a more comprehensive understanding of the evolution of polycentric structures and spatial effects [41].…”

Reward shaping using directed graph convolution neural networks for reinforcement learning and games

“…Over the past few decades, game theory has utilized the concepts and methods of reinforcement learning (RL) to solve decision-making problems [1]; [2]; [3]. An RL problem can be seen as a game between individual decision-makers and the environment [4]; [5].…”