Weighted automata extraction and explanation of recurrent neural networks for natural language tasks

Journal article

Recurrent Neural Networks (RNNs) have achieved tremendous success in processing sequential data, yet understanding and analyzing their behaviours remains a significant challenge. To this end, many efforts have been made to extract finite automata from RNNs, which are more amenable for analysis and explanation. However, existing approaches like exact learning and compositional approaches for model extraction have limitations in either scalability or precision. In this paper, we propose a novel framework of Weighted Finite Automata (WFA) extraction and explanation to tackle the limitations for natural language tasks. First, to address the transition sparsity and context loss problems we identified in WFA extraction for natural language tasks, we propose an empirical method to complement missing rules in the transition diagram, and adjust transition matrices to enhance the context-awareness of the WFA. We also propose two data augmentation tactics to track more dynamic behaviours of RNN, which further allows us to improve the extraction precision. Based on the extracted model, we propose an explanation method for RNNs including a word embedding method – Transition Matrix Embeddings (TME) and TME-based task oriented explanation for the target RNN. Our evaluation demonstrates the advantage of our method in extraction precision than existing approaches, and the effectiveness of TME-based explanation method in applications to pretraining and adversarial example generation.

Authors: Wei, Z; Zhang, X; Zhang, Y; Sun, M

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Elsevier
• Journal: Journal of Logical and Algebraic Methods in Programming
• Volume: 136
• Article number: 100907
• Publication date: 2023-09-06
• Acceptance date: 2023-08-31
• DOI: 10.1016/j.jlamp.2023.100907
• EISSN: 2352-2216
• ISSN: 2352-2208
• Language: English
• Keywords: abstraction; recurrent neural networks; weighted finite automata; explanation; natural languages