Bias-generating Agent-based Simulation and its Application to Election Systems

Agent-based-Simulation can be used to study the impact of individual decisions on the overall outcome. There has been a lot of research focused on giving the agent the ability to learn, especially there are studies that analyze corporate organizations using reinforcement learning. We focus on the election phenomena that there are often cases where popular candidates are suddenly overtaken by others in the election campaign. We try to use the characteristic of overfitting of neural networks and conditional reflex learning, such as the “Pavlov’s dog” effect, to explain the phenomena.