Knowledge Graph Embedding Circuitry Predicts Surprising Facts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing prediction models for knowledge graphs fail to account for surprise when generating predictions, resulting in generic outcomes, and lack the ability to discover new, unforeseen facts, which are desirable in applications like AI and formulation fields.
Innovation Solution
The implementation of a computing device with reception, embedding, training, inference, and processing circuitry that converts knowledge graphs into an embeddings space, applies relational learning techniques, and calculates surprise scores to identify new surprising facts through geometric and surprise score functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional prediction models are used for knowledge graphs, then the system is simple and easy to implement, but the predictions are generic and fail to discover new surprising facts
Solution Approach 1:
The patent transforms knowledge graph data into an embedding space, adding a dimensional representation layer that enables the system to capture complex relationships and discover surprising facts that traditional flat prediction models cannot detect
Solution Approach 2:
The embedding space acts as an intermediary between the knowledge graph structure and the prediction output, allowing the system to infer new facts through relational learning while maintaining the original graph's semantic meaning
2Measurement precision
If relational learning techniques are applied to discover new facts, then the accuracy and relevance of predictions improve, but the computational resources and processing time increase
Solution Approach 1:
The system performs preliminary embedding transformation and training before actual prediction, pre-computing relational representations that accelerate subsequent inference and reduce real-time computational requirements
Solution Approach 2:
The patent changes the representation parameters of knowledge graph entities into embedding vectors, enabling efficient relational learning through mathematical operations on these transformed parameters rather than direct graph traversal
Data Source
AI summary
Complex computer system architectures are described for analyzing data elements of a knowledge graph, and predicting new surprising or unforeseen facts from relational learning applied to the knowledge graph. This discovery process takes advantage of the knowledge graph structure to improve the computing capabilities of a device executing a discovery calculation by applying both training and inference analysis techniques on the knowledge graph within an embedding space, and generating a scoring strategy for predicting surprising facts that may be discoverable from the knowledge graph.