English

State-Regularized Recurrent Neural Networks to Extract Automata and Explain Predictions

Machine Learning 2022-12-13 v1

Abstract

Recurrent neural networks are a widely used class of neural architectures. They have, however, two shortcomings. First, they are often treated as black-box models and as such it is difficult to understand what exactly they learn as well as how they arrive at a particular prediction. Second, they tend to work poorly on sequences requiring long-term memorization, despite having this capacity in principle. We aim to address both shortcomings with a class of recurrent networks that use a stochastic state transition mechanism between cell applications. This mechanism, which we term state-regularization, makes RNNs transition between a finite set of learnable states. We evaluate state-regularized RNNs on (1) regular languages for the purpose of automata extraction; (2) non-regular languages such as balanced parentheses and palindromes where external memory is required; and (3) real-word sequence learning tasks for sentiment analysis, visual object recognition and text categorisation. We show that state-regularization (a) simplifies the extraction of finite state automata that display an RNN's state transition dynamic; (b) forces RNNs to operate more like automata with external memory and less like finite state machines, which potentiality leads to a more structural memory; (c) leads to better interpretability and explainability of RNNs by leveraging the probabilistic finite state transition mechanism over time steps.

Keywords

Cite

@article{arxiv.2212.05178,
  title  = {State-Regularized Recurrent Neural Networks to Extract Automata and Explain Predictions},
  author = {Cheng Wang and Carolin Lawrence and Mathias Niepert},
  journal= {arXiv preprint arXiv:2212.05178},
  year   = {2022}
}

Comments

To appear at IEEE Transactions on Pattern Analysis and Machine Intelligence. The extended version of State-Regularized Recurrent Neural Networks [arXiv:1901.08817]

R2 v1 2026-06-28T07:28:41.462Z