Knowledge Graph Embeddings and Explainable AI

Abstract

Knowledge graph embeddings are now a widely adopted approach to knowledge representation in which entities and relationships are embedded in vector spaces. In this chapter, we introduce the reader to the concept of knowledge graph embeddings by explaining what they are, how they can be generated and how they can be evaluated. We summarize the state-of-the-art in this field by describing the approaches that have been introduced to represent knowledge in the vector space. In relation to knowledge representation, we consider the problem of explainability, and discuss models and methods for explaining predictions obtained via knowledge graph embeddings.

Federico Bianchi, Gaetano Rossiello, Luca Costabello, Matteo Plamonari, Pasquale Minervini, Knowledge Graph Embeddings and Explainable AI. In: Ilaria Tiddi, Freddy Lecue, Pascal Hitzler (eds.), Knowledge Graphs for eXplainable AI -- Foundations, Applications and Challenges. Studies on the Semantic Web, IOS Press, Amsterdam, 2020

References in corpus (15)

• Complex Embeddings for Simple Link Prediction
• GNNExplainer: Generating Explanations for Graph Neural Networks
• SimplE Embedding for Link Prediction in Knowledge Graphs
• Analogical Inference for Multi-Relational Embeddings
• PyTorch-BigGraph: A Large-scale Graph Embedding System
• Representation Tradeoffs for Hyperbolic Embeddings
• DKN: Deep Knowledge-Aware Network for News Recommendation
• Poincaré GloVe: Hyperbolic Word Embeddings
• Logic Tensor Networks: Deep Learning and Logical Reasoning from Data and Knowledge
• Adversarial Sets for Regularising Neural Link Predictors
• Hyperbolic Disk Embeddings for Directed Acyclic Graphs
• HyperKG: Hyperbolic Knowledge Graph Embeddings for Knowledge Base Completion
• Probability Calibration for Knowledge Graph Embedding Models
• Entity Embeddings with Conceptual Subspaces as a Basis for Plausible Reasoning
• Recurrent One-Hop Predictions for Reasoning over Knowledge Graphs