Machine learning phases in swarming systems

Abstract Recent years have witnessed a growing interest in using machine learning to predict and identify phase transitions (PTs) various systems. Here we adopt convolutional neural networks (CNNs) study the PTs of Vicsek model, solving problem that traditional order parameters are insufficiently able do. Within large-scale simulations, there four phases, confirm all between two neighboring phases first-order. We successfully classified by CNNs with high accuracy identified PT points, while approaches fail obtain. These results indicate great potential approach understanding complexities collective behaviors, related complex systems general.