LATENTLOGIC: Learning Logic Rules in Latent Space over Knowledge Graphs

Junnan Liu, Qianren Mao, Chenghua Lin, Yangqiu Song, Jianxin Li


Abstract
Learning logic rules for knowledge graph reasoning is essential as such rules provide interpretable explanations for reasoning and can be generalized to different domains. However, existing methods often face challenges such as searching in a vast search space (e.g., enumeration of relational paths or multiplication of high-dimensional matrices) and inefficient optimization (e.g., techniques based on reinforcement learning or EM algorithm). To address these limitations, this paper proposes a novel framework called LatentLogic to efficiently mine logic rules by controllable generation in the latent space. Specifically, to map the discrete relational paths into the latent space, we leverage a pre-trained VAE and employ a discriminator to establish an energy-based distribution. Additionally, we incorporate a sampler based on ordinary differential equations, enabling the efficient generation of logic rules in our approach. Extensive experiments on benchmark datasets demonstrate the effectiveness and efficiency of our proposed method.
Anthology ID:
2023.findings-emnlp.304
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2023
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4578–4586
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.304
DOI:
10.18653/v1/2023.findings-emnlp.304
Bibkey:
Cite (ACL):
Junnan Liu, Qianren Mao, Chenghua Lin, Yangqiu Song, and Jianxin Li. 2023. LATENTLOGIC: Learning Logic Rules in Latent Space over Knowledge Graphs. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 4578–4586, Singapore. Association for Computational Linguistics.
Cite (Informal):
LATENTLOGIC: Learning Logic Rules in Latent Space over Knowledge Graphs (Liu et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-emnlp.304.pdf