Paper

Augmentic Compositional Models for Knowledge Base Completion Using Gradient Representations

---

Abstract

Neural models of Knowledge Base data have typically employed compositional representations of graph objects: entity and relation embeddings are systematically combined to evaluate the truth of a candidate Knowedge Base entry. Using a model inspired by Harmonic Grammar, we propose to tokenize triplet embeddings by subjecting them to a process of optimization with respect to learned well-formedness conditions on Knowledge Base triplets. The resulting model, known as Gradient Graphs, leads to sizable improvements when implemented as a companion to compositional models. Also, we show that the "supracompositional" triplet token embeddings it produces have interpretable properties that prove helpful in performing inference on the resulting triplet representations.

Keywords: Computational Semantics, Knowledge Base Completion, Neural Networks

How to Cite:

Lalisse, M. R. & Smolensky, P., (2019) “Augmentic Compositional Models for Knowledge Base Completion Using Gradient Representations”, Society for Computation in Linguistics 2(1), 257-266. doi: https://doi.org/10.7275/8et8-qd83