Learning grammar with a divide-and-concur neural network

TL;DR

This paper introduces a divide-and-concur neural network approach for inferring context-free grammars that is interpretable, data-efficient, and capable of refining or expanding existing grammars from limited data.

Contribution

It presents a novel iterative projection method for grammar inference that requires fewer parameters and less data than traditional models, enabling interpretability and flexibility.

Findings

Successfully infers grammars from few sentences

Can refine existing grammars and expand lexicons

Produces interpretable grammatical rules

Abstract

We implement a divide-and-concur iterative projection approach to context-free grammar inference. Unlike most state-of-the-art models of natural language processing, our method requires a relatively small number of discrete parameters, making the inferred grammar directly interpretable -- one can read off from a solution how to construct grammatically valid sentences. Another advantage of our approach is the ability to infer meaningful grammatical rules from just a few sentences, compared to the hundreds of gigabytes of training data many other models employ. We demonstrate several ways of applying our approach: classifying words and inferring a grammar from scratch, taking an existing grammar and refining its categories and rules, and taking an existing grammar and expanding its lexicon as it encounters new words in new data.