AAAI 2025technical0 citations
An Evaluation of Approaches to Train Embeddings for Logical Inference (Student Abstract)
Yasir White, Jevon Lipsey, Jeff Heflin
Abstract
Knowledge bases traditionally require manual optimization to ensure reasonable performance when answering queries. We build on previous neurosymbolic approaches by improving the training of an embedding model for logical statements that maximizes similarity between unifying atoms and minimizes similarity of non-unifying atoms. In particular, we evaluate different approaches to training this model.
BibTeX
@article{White_Lipsey_Heflin_2025, title={An Evaluation of Approaches to Train Embeddings for Logical Inference (Student Abstract)}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/35313}, DOI={10.1609/aaai.v39i28.35313}, abstractNote={Knowledge bases traditionally require manual optimization to ensure reasonable performance when answering queries. We build on previous neurosymbolic approaches by improving the training of an embedding model for logical statements that maximizes similarity between unifying atoms and minimizes similarity of non-unifying atoms. In particular, we evaluate different approaches to training this model.}, number={28}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={White, Yasir and Lipsey, Jevon and Heflin, Jeff}, year={2025}, month={Apr.}, pages={29527-29528} }