Deep neural network system for similarity-based graph representations

A method of determining graph similarity, the method comprising:

• receiving, by one or more processors, a first graph;
• receiving, by the one or more processors, a second graph;
• generating, by one or more neural networks, a vector representation of the first graph;
• generating, by the one or more neural networks, a vector representation of the second graph, wherein generating, by the one or more neural networks, a vector representation of a graph comprises:
• processing, by the one or more neural networks, the graph to generate a node state representation vector for each node of the graph and an edge representation vector for each edge of the graph;
• and
• processing, by the one or more neural networks, the node state representation vectors and the edge representation vectors to generate the vector representation of the graph;
• determining, by the one or more processors, a similarity score based upon the vector representations of the first graph and the second graph, wherein the one or more neural networks are configured to process a pair of input graphs to generate a vector representation of each input graph of the pair of input graphs, the vector representation of each input graph being based upon both input graphs of the pair of graphs;
• and wherein
• the vector representation of the first graph and the vector representation of the second graph are generated based upon inputting the first graph and the second graph as a pair of input graphs to the one or more neural networks.

One or more non-transitory computer-readable storage media storing instructions that when executed by one or more computers cause the one or more computers to implement a neural network system for determining graph similarity, the neural network system comprising:

• one or more neural networks configured to:
• process an input graph to generate a node state representation vector for each node of the input graph and an edge representation vector for each edge of the input graph;
• and
• process the node state representation vectors and the edge representation vectors to generate a vector representation of the input graph;
• one or more processors configured to:
• receive a first graph;
• receive a second graph;
• generate a vector representation of the first graph using the one or more neural networks;
• generate a vector representation of the second graph using the one or more neural networks;
• determine a similarity score for the first graph and the second graph based upon the vector representations of the first graph and the second graph, wherein the one or more neural networks are configured to process a pair of input graphs to generate a vector representation of each input graph of the pair of input graphs, the vector representation of each input graph being based upon both input graphs of the pair of graphs;
• and wherein
• the vector representation of the first graph and the vector representation of the second graph are generated based upon inputting the first graph and the second graph as a pair of input graphs to the one or more neural networks.