Patent · US Active

Interpretable visualization system for graph neural network

US12182698B2 · kind B2 · utility

0Cited by

2References

14Claims

0Family size

Assignees

Inventors

Dakuo Wang · Cambridge, US
Sijia Liu · Beijing, CN
Abel Valente · Dieciséis de Julio, AR
Chuang Gan · Cambridge, US
Bei Chen · Dublin, IE
Dongyu Liu · Fuzhou, CN
Yi Sun · Sha Tin, CN

Key dates

Filing date	Sep 30, 2020
Grant date	Dec 31, 2024
Priority date	—
Expiry date	Oct 12, 2043

Classification

Technology area (CPC G)Physics
CPC primaryG06N3/09
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Use a computerized trained graph neural network model to classify an input instance with a predicted label. With a computerized graph neural network interpretation module, compute a gradient-based saliency matrix based on the input instance and the predicted label, by taking a partial derivative of class prediction with respect to an adjacency matrix of the model. With a computerized user interface, obtain user input responsive to the gradient-based saliency matrix. Optionally, modify the trained graph neural network model based on the user input; and re-classify the input instance with a new predicted label based on the modified trained graph neural network model.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.