Graph neural network: Difference between revisions

Graph neural network (view source)

960 bytes added , 10 June 2022

no edit summary

5,336

edits

@@ Line 1: / Line 1: @@
+If you can represent your data as a graph, you can use a graph neural network to perform inference on the data and output a new graph or predictions on the graph.
+==Introduction==
+===Structure===
+A graph neural network consists of layers which operate on graphs.
+At each layer, you get the following features during inference:
+* The graph structure
+* Node embeddings
+* Edge embeddings
+* Graph embeddings
+===Message Passing Layer===
+A standard GNN layer consists of pooling functions followed by update functions.
+===Pooling===
+There are many types of pooling to choose from:
+* For nodes, you can add the node embedding to connected node embedding or connected edge embeddings.
+* Similarly for edges, you can add its embedding to connected node embeddings.
+* For the entire graph, you can add all node embeddings together.
+After pooling, you can use update function (e.g. MLP) to update the embeddings/state at each node, edge, and for the entire graph.
 ==Resources==
 * [https://distill.pub/2021/gnn-intro/ A Gentle Introduction to Graph Neural Networks (2021) by Google Research]
 * [https://distill.pub/2021/understanding-gnns/ Understanding Convolutions on Graphs (2021) by Google Research]