Revisiting Simple Neural Networks for Learning Representations of Knowledge Graphs
This work addresses knowledge base completion for AI applications, but it is incremental as it revisits simple methods rather than introducing a new paradigm.
The paper tackles the problem of learning vector representations for knowledge graphs to complete missing triples, showing that a simple neural network score function achieves near state-of-the-art performance consistently across multiple datasets, while also identifying biases in standard benchmarks.
We address the problem of learning vector representations for entities and relations in Knowledge Graphs (KGs) for Knowledge Base Completion (KBC). This problem has received significant attention in the past few years and multiple methods have been proposed. Most of the existing methods in the literature use a predefined characteristic scoring function for evaluating the correctness of KG triples. These scoring functions distinguish correct triples (high score) from incorrect ones (low score). However, their performance vary across different datasets. In this work, we demonstrate that a simple neural network based score function can consistently achieve near start-of-the-art performance on multiple datasets. We also quantitatively demonstrate biases in standard benchmark datasets, and highlight the need to perform evaluation spanning various datasets.