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

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 e...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Journal of logical and algebraic methods in programming Ročník 136; s. 100907
Hlavní autori: Wei, Zeming, Zhang, Xiyue, Zhang, Yihao, Sun, Meng
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Inc 01.01.2024
Predmet:
Abstraction
Explanation
Natural languages
Recurrent neural networks
Weighted finite automata
Weighted finite automata
Recurrent neural networks
Abstraction
Natural languages
Explanation
ISSN:2352-2208
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí: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.
ISSN:2352-2208
DOI:10.1016/j.jlamp.2023.100907