Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies

Abstract

The success of long short-term memory (LSTM) neural networks in language processing is typically attributed to their ability to capture long-distance statistical regularities. Linguistic regularities are often sensitive to syntactic structure; can such dependencies be captured by LSTMs, which do not have explicit structural representations? We begin addressing this question using number agreement in English subject-verb dependencies. We probe the architecture's grammatical competence both using training objectives with an explicit grammatical target (number prediction, grammaticality judgments) and using language models. In the strongly supervised settings, the LSTM achieved very high overall accuracy (less than 1% errors), but errors increased when sequential and structural information conflicted. The frequency of such errors rose sharply in the language-modeling setting. We conclude that LSTMs can capture a non-trivial amount of grammatical structure given targeted supervision, but stronger architectures may be required to further reduce errors; furthermore, the language modeling signal is insufficient for capturing syntax-sensitive dependencies, and should be supplemented with more direct supervision if such dependencies need to be captured.

arXiv comment15 pages; to appear in Transactions of the Association for Computational Linguistics

References in corpus (6)

• Exploring the Limits of Language Modeling
• Visualizing and Understanding Recurrent Networks
• Inferring Algorithmic Patterns with Stack-Augmented Recurrent Nets
• Learning to Transduce with Unbounded Memory
• Tree-structured composition in neural networks without tree-structured architectures
• Representation of linguistic form and function in recurrent neural networks

Cited by in corpus (9)

• Land Cover Classification via Multi-temporal Spatial Data by Recurrent Neural Networks
• Theoretical Limitations of Self-Attention in Neural Sequence Models
• Interpreting and improving natural-language processing (in machines) with natural language-processing (in the brain)
• Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
• Understanding Early Word Learning in Situated Artificial Agents
• Augmenting Neural Machine Translation with Knowledge Graphs
• Inducing Regular Grammars Using Recurrent Neural Networks
• Learning Noun Cases Using Sequential Neural Networks
• MTLHealth: A Deep Learning System for Detecting Disturbing Content in Student Essays