Today, the dominant paradigm for training neural networks involves minimizing task loss on a large dataset. Using world knowledge to inform a model, and yet retain the ability to perform end-to-end training remains an open question. In this paper, we present a novel framework for introducing declarative knowledge to neural network architectures in order to guide training and prediction. Our framework systematically compiles logical statements into computation graphs that augment a neural network without extra learnable parameters or manual redesign. We evaluate our modeling strategy on three tasks: machine comprehension, natural language inference, and text chunking. Our experiments show that knowledge-augmented networks can strongly improve over baselines, especially in low-data regimes.
@article{arxiv.1906.06298,
title = {Augmenting Neural Networks with First-order Logic},
author = {Tao Li and Vivek Srikumar},
journal= {arXiv preprint arXiv:1906.06298},
year = {2020}
}
Comments
Accepted in ACL 2019. Minor fixes in Fig 4; extra citation in related works; Typo fix in constraint N3 and its description