search

SHTUPLUS / PySGG

The toolkit for scene graph generation
Other
71 stars 15 forks source link

The question about message propagation #2

Closed rlqja1107 closed 2 years ago

rlqja1107 commented 3 years ago

Above picture describes the message propagation method in GCN(Graph Convolution Network). In this method, when the information of the neighborhood is aggregated, it adds the message itself, then makes a transformation and adopts the activation function.

However, the BGNN method shows that for updating the node representation, the l-th representation of node is added after the message of neighborhood is transformed and is adopted the activation function.
I wonder why BGNN takes the latter method.

Scarecrow0 commented 3 years ago

Hi!

Thanks for attending our paper.

Since there are many graph iterations in our forward time (N graph iteration in K stage refinement), which leads to a quite deep GNN. To this end, we introduce the skip-connection mechanism into our GNN design, as shown in our formula.