What model architectures does Kumo incorporate into its GNN design search space?

Kumo recommendations are powered by a GNN architecture, inspired by several academic papers in recent history. Data scientists can benefit from these advances in model architecture, without needing to code them up manually.

The following are some research projects used by Kumo:

• GraphSAGE does inductive representation learning to deliver great recommendations for users with very little interaction data such as first or second time visitors
• ID-GNN is a training process that enables the model to learn patterns such as repeat purchase or brand affinity
• PNA introduces a variety of aggregation operators which are explored by Kumo AutoML
• GCN describes mean-pooling aggregation, which captures similarity between users with similar item purchases
• GIN captures frequency signal to learn more complex user behavior like power users vs resurrected users
• NBF networks reduce the computational cost of models, by providing an efficient way to capture paths between nodes
• GraphMixer uses temporal representation learning, to interpret sequences of user actions such as on-site browsing history
• RDL introduces temporal sampling, which learns from past sequences of user actions to predict the future.

As a result of this hybrid GNN approach, Kumo is able to achieve great predictive performance out of the box, across a wide range of challenging scenarios.