A Solution to Neural Message Passing
Wu Dai, Jinyang Wang, Tianyi He, Chunyan Liu · 2020 International Conference on Computer Engineering and Application (ICCEA) · 2020
Because of the learning process of the messaging neural network from the inside out, it can explain a game almost perfectly, from the micro to the macro. We need to feed it a certain amount of data. We set a half court as the unit for our AI. Every machine, even AI, is born to work. After our message passing neural network has learned to recurse and reasoning, it is time to get to work. With trained AI, we can explore unknown data spaces. All we need to do is to change the location properties of a few nodes on the input multidimensional graph. We change the structural strategy of all the data sample graph, then we can calculate the average performance indicators of this structural strategy.