Structure Aware Negative Sampling in Knowledge Graphs

Kian Ahrabian, Aarash Feizi, Yasmin Salehi, William L. Hamilton, Avishek Joey Bose


Abstract
Learning low-dimensional representations for entities and relations in knowledge graphs using contrastive estimation represents a scalable and effective method for inferring connectivity patterns. A crucial aspect of contrastive learning approaches is the choice of corruption distribution that generates hard negative samples, which force the embedding model to learn discriminative representations and find critical characteristics of observed data. While earlier methods either employ too simple corruption distributions, i.e. uniform, yielding easy uninformative negatives or sophisticated adversarial distributions with challenging optimization schemes, they do not explicitly incorporate known graph structure resulting in suboptimal negatives. In this paper, we propose Structure Aware Negative Sampling (SANS), an inexpensive negative sampling strategy that utilizes the rich graph structure by selecting negative samples from a node’s k-hop neighborhood. Empirically, we demonstrate that SANS finds semantically meaningful negatives and is competitive with SOTA approaches while requires no additional parameters nor difficult adversarial optimization.
Anthology ID:
2020.emnlp-main.492
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6093–6101
Language:
URL:
https://aclanthology.org/2020.emnlp-main.492
DOI:
10.18653/v1/2020.emnlp-main.492
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2020.emnlp-main.492.pdf
Optional supplementary material:
 2020.emnlp-main.492.OptionalSupplementaryMaterial.zip
Video:
 https://slideslive.com/38938925
Code
 kahrabian/SANS