Convolutional Complex Knowledge Graph Embeddings

Caglar Demir, Axel Ngonga

Keywords: complex knowledge graph embeddings, convolutions

Abstract: In this paper, we study the problem of learning continuous vector representations of knowledge graphs for predicting missing links. We present a new approach called ConEx, which infers missing links by leveraging the composition of a 2D convolution with a Hermitian inner product of complex-valued embedding vectors. We evaluate ConEx against state-of-the-art approaches on the WN18RR, FB15K-237, KINSHIP and UMLS benchmark datasets. Our experimental results show that ConEx achieves a performance superior to that of state-of-the-art approaches such as RotatE, QuatE and TuckER in the link prediction task on all datasets while requiring at least 8 times fewer parameters. We ensure the reproducibility of our results by providing an open-source implementation which includes the training, evaluation scripts along with pre-trained models at https://github.com/conex-kge/ConEx.

One-sentence Summary: We propose a new knowledge graph embedding approach combining a 2D convolution with an Hermitian inner product of complex-valued embeddings to tackle the link prediction problem.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 3 code implementations](https://www.catalyzex.com/paper/arxiv:2008.03130/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=fHwFb2b0aV