On the Practical Ability of Recurrent Neural Networks to Recognize Hierarchical Languages

TL;DR

This paper investigates the ability of recurrent neural networks to recognize hierarchical structures in languages, revealing they generalize well within bounded depths but struggle with longer sequences, which explains their effectiveness in natural language processing.

Contribution

The study demonstrates that recurrent models can recognize bounded-depth Dyck languages and generalize to longer sequences within those bounds, clarifying their success in hierarchical language modeling.

Findings

Recurrent models perform well on same-range training and test lengths.

They struggle with longer test sequences unless bounded depth is considered.

Transformers show different generalization behaviors.

Abstract

While recurrent models have been effective in NLP tasks, their performance on context-free languages (CFLs) has been found to be quite weak. Given that CFLs are believed to capture important phenomena such as hierarchical structure in natural languages, this discrepancy in performance calls for an explanation. We study the performance of recurrent models on Dyck-n languages, a particularly important and well-studied class of CFLs. We find that while recurrent models generalize nearly perfectly if the lengths of the training and test strings are from the same range, they perform poorly if the test strings are longer. At the same time, we observe that recurrent models are expressive enough to recognize Dyck words of arbitrary lengths in finite precision if their depths are bounded. Hence, we evaluate our models on samples generated from Dyck languages with bounded depth and find that they are indeed able to generalize to much higher lengths. Since natural language datasets have nested dependencies of bounded depth, this may help explain why they perform well in modeling hierarchical dependencies in natural language data despite prior works indicating poor generalization performance on Dyck languages. We perform probing studies to support our results and provide comparisons with Transformers.