Neural Network Model Of Message Passing in Team Cooperation

The football match is so wonderful that every moment on the pitch touch the hearts of the spectators. What if two players are in contact on the pitch? To this end, we will build a neural network model to more intuitively reflect the players' interactions in the game by observing the small interactions between the players in the team, thus can more clearly reflect the entire competition condition. The model can model dynamic information as the graph develops. In particular, the proposed framework can update the node information by continuously capturing the sequential information of edges, the time interval between edges, and the information propagation. Experimental results on various dynamic graphs demonstrate the effectiveness of the proposed framework.