A Soft Version of Predicate Invention Based on Structured Sparsity

Abstract

In predicate invention (PI), new predicates are introduced into a logical theory, usually by rewriting a group of closely-related rules to use a common invented predicate as a "subroutine". PI is difficult, since a poorly-chosen invented predicate may lead to error cascades. Here we suggest a "soft" version of predicate invention: instead of explicitly creating new predicates, we implicitly group closely-related rules by using structured sparsity to regularize their parameters together. We show that soft PI, unlike hard PI, consistently improves over previous strong baselines for structure-learning on two large-scale tasks.

Cite

Text

Wang et al. "A Soft Version of Predicate Invention Based on Structured Sparsity." International Joint Conference on Artificial Intelligence, 2015.

Markdown

[Wang et al. "A Soft Version of Predicate Invention Based on Structured Sparsity." International Joint Conference on Artificial Intelligence, 2015.](https://mlanthology.org/ijcai/2015/wang2015ijcai-soft/)

BibTeX

@inproceedings{wang2015ijcai-soft,
  title     = {{A Soft Version of Predicate Invention Based on Structured Sparsity}},
  author    = {Wang, William Yang and Mazaitis, Kathryn and Cohen, William W.},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2015},
  pages     = {3918-3924},
  url       = {https://mlanthology.org/ijcai/2015/wang2015ijcai-soft/}
}