# How to estimate the convolutional representation of a graph from its similarity to other graph convolutional representation?

Suppose we have two graphs A and B disconnected to each other (let's say 2-hops each), within a larger graph. If the convolutional representation of graph A is known, is it possible to estimate the definitive convolutional representation of graph B based on its similarity to graph A?

If yes, what do you think is the simplest (arithmetically) way to do this, which algorithm can help me to do this? You can assume that precision requirements are not important.