Scalable Knowledge Refactoring Using Constrained Optimisation
Abstract
Knowledge refactoring compresses logic programs by replacing them with new rules. Current approaches struggle to scale to large programs. To overcome this limitation, we introduce a constrained optimisation refactoring approach. Our first key idea is to encode the problem with decision variables based on literals rather than rules. Our second key idea is to focus on linear invented rules. Our empirical results on multiple domains show that our approach can refactor programs quicker and with more compression than the previous state-of-the-art approach, sometimes by 60%.
Cite
Text
Liu et al. "Scalable Knowledge Refactoring Using Constrained Optimisation." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I14.33650Markdown
[Liu et al. "Scalable Knowledge Refactoring Using Constrained Optimisation." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/liu2025aaai-scalable/) doi:10.1609/AAAI.V39I14.33650BibTeX
@inproceedings{liu2025aaai-scalable,
title = {{Scalable Knowledge Refactoring Using Constrained Optimisation}},
author = {Liu, Minghao and Cerna, David M. and Gouveia, Filipe and Cropper, Andrew},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2025},
pages = {15049-15057},
doi = {10.1609/AAAI.V39I14.33650},
url = {https://mlanthology.org/aaai/2025/liu2025aaai-scalable/}
}