17819248. SYSTEMS AND METHODS FOR GENERATING MULTIPURPOSE GRAPH NODE EMBEDDINGS FOR MACHINE LEARNING simplified abstract (Capital One Services, LLC)

From WikiPatents
Jump to navigation Jump to search

SYSTEMS AND METHODS FOR GENERATING MULTIPURPOSE GRAPH NODE EMBEDDINGS FOR MACHINE LEARNING

Organization Name

Capital One Services, LLC

Inventor(s)

Jiankun Liu of Flower Mound TX (US)

Aamer Charania of Flower Mound TX (US)

Behrouz Saghafi Khadem of Frisco TX (US)

SYSTEMS AND METHODS FOR GENERATING MULTIPURPOSE GRAPH NODE EMBEDDINGS FOR MACHINE LEARNING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17819248 titled 'SYSTEMS AND METHODS FOR GENERATING MULTIPURPOSE GRAPH NODE EMBEDDINGS FOR MACHINE LEARNING

Simplified Explanation

In some aspects, a computing system may create different node embeddings (e.g., different instances of the same graph) and aggregate the node embeddings to form “multipurpose” node embeddings. As an example, the different node embeddings may include node embeddings generated using unsupervised machine learning and node embeddings generated using supervised machine learning. In this way, for example, a variety of machine learning models may use the aggregated node embeddings without the need for each machine learning model to generate separate node embeddings each time a machine learning task is performed. A machine learning model may use all or a portion of the features in the aggregated node embeddings as appropriate for the task the model is performing.

  • The computing system creates different node embeddings and aggregates them to form multipurpose node embeddings.
  • Node embeddings can be generated using unsupervised and supervised machine learning techniques.
  • Machine learning models can utilize the aggregated node embeddings without needing to generate separate embeddings for each task.
  • Models can use all or a portion of the features in the aggregated node embeddings based on the task at hand.

Potential Applications

  • Machine learning tasks that require multiple types of node embeddings.
  • Graph-based applications that benefit from multipurpose node embeddings.
  • Recommendation systems that utilize different types of node embeddings for improved accuracy.

Problems Solved

  • Reducing the need to generate separate node embeddings for different machine learning tasks.
  • Simplifying the process of utilizing node embeddings in various machine learning models.
  • Enhancing the efficiency and effectiveness of machine learning algorithms that rely on node embeddings.

Benefits

  • Increased efficiency in utilizing node embeddings across different machine learning models.
  • Improved performance of machine learning tasks by leveraging multipurpose node embeddings.
  • Simplification of the machine learning process by aggregating and using different types of node embeddings.


Original Abstract Submitted

In some aspects, a computing system may create different node embeddings (e.g., different instances of the same graph) and aggregate the node embeddings to form “multipurpose” node embeddings. As an example, the different node embeddings may include node embeddings generated using unsupervised machine learning and node embeddings generated using supervised machine learning. In this way, for example, a variety of machine learning models may use the aggregated node embeddings without the need for each machine learning model to generate separate node embeddings each time a machine learning task is performed. A machine learning model may use all or a portion of the features in the aggregated node embeddings as appropriate for the task the model is performing.

Retrieved from "http://wikipatents.org/index.php?title=17819248._SYSTEMS_AND_METHODS_FOR_GENERATING_MULTIPURPOSE_GRAPH_NODE_EMBEDDINGS_FOR_MACHINE_LEARNING_simplified_abstract_(Capital_One_Services,_LLC)&oldid=35767"