US 12,488,238 B2
Information-aware graph contrastive learning
Wei Cheng, Princeton Junction, NJ (US); Dongkuan Xu, State College, PA (US); and Haifeng Chen, West Windsor, NJ (US)
Assigned to NEC Corporation, Tokyo (JP)
Filed by NEC Laboratories America, Inc., Princeton, NJ (US)
Filed on Apr. 25, 2022, as Appl. No. 17/728,071.
Claims priority of provisional application 63/316,505, filed on Mar. 4, 2022.
Claims priority of provisional application 63/191,367, filed on May 21, 2021.
Prior Publication US 2022/0383108 A1, Dec. 1, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/08 (2023.01)
CPC G06N 3/08 (2013.01) 20 Claims
OG exemplary drawing
 
1. A method for performing contrastive learning for graph tasks and datasets by employing an information-aware graph contrastive learning framework, the method comprising:
obtaining two semantically similar views of a graph coupled with a label for training by employing a view augmentation component to generate views of the graph that do not affect a semantic label of the graph;
feeding the two semantically similar views into respective separate instances of encoder networks to extract latent representations preserving both structure and attribute information in the two semantically similar views;
selecting a contrastive mode that is characterized by an aggregation operation by solving an optimization problem:

OG Complex Work Unit Math
where c*i and c*j are aggregation operations applied to latent representations z*i and z*j and where I(.;.) is a mutual information;
minimizing a contrastive loss based on the contrastive mode by maximizing feature consistency between the latent representations;
training a neural network by performing graph contrastive learning with the minimized contrastive loss; and
predicting a new graph label or a new node label in the graph using the neural network trained by the graph contrastive learning, wherein the predictive new graph label or new node label is used to support a decision making process.