System and method for structure learning for graph neural networks
AssigneeRoyal Bank of Canada
InventorsBahare FATEMI, Seyed Mehran KAZEMI, Layla EL ASRI
Life of this patent
Estimating the term…
Abstract
A graph structure having nodes and edges is represented as an adjacency matrix, and nodes of the graph structure have node features. A computer-implemented method and system for generating a graph structure are provided, the method comprising: generating an adjacency matrix based on a plurality of node features; generating a plurality of noisy node features based on the plurality of node features; generating a plurality of denoised node features using a neural network based on the plurality of noisy node features and the adjacency matrix; and updating the adjacency matrix based on the plurality of denoised node features.
Claims25
A computer-implemented system for generating a graph structure represented as an adjacency matrix, the system comprising:
- at least one processor;
- and
- memory in communication with said at least one processor, wherein the memory stores instructions, when executed at said at least one processor, causes said system to:
- generate an adjacency matrix based on a plurality of node features;
- generate a plurality of noisy node features based on the plurality of node features;
- generate a plurality of denoised node features using a first neural network based on the plurality of noisy node features and the adjacency matrix;
- and
- update the adjacency matrix based on the plurality of denoised node features.
A computer-implemented method for generating a graph structure represented as an adjacency matrix, the method comprising:
- generating an adjacency matrix based on a plurality of node features;
- generating a plurality of noisy node features based on the plurality of node features;
- generating a plurality of denoised node features using a first neural network based on the plurality of noisy node features and the adjacency matrix;
- and
- updating the adjacency matrix based on the plurality of denoised node features.
A non-transitory computer-readable storage medium storing a data model representing a neural network for generating a graph structure, wherein the neural network is trained by:
- generating an adjacency matrix based on a plurality of node features;
- generating a plurality of noisy node features based on the plurality of node features;
- generating a plurality of denoised node features using the neural network based on the plurality of noisy node features and the adjacency matrix;
- and
- updating the adjacency matrix based on the plurality of denoised node features.
Dependencies are inferred from the claim text.