Extracting Weighted Finite Automata from Recurrent Neural Networks for Natural Languages. (arXiv:2206.14621v2 [cs.CL] UPDATED)

Recurrent Neural Networks (RNNs) have achieved tremendous success in
sequential data processing. However, it is quite challenging to interpret and
verify RNNs' behaviors directly. To this end, many efforts have been made to
extract finite automata from RNNs. Existing approaches such as exact learning
are effective in extracting finite-state models to characterize the state
dynamics of RNNs for formal languages, but are limited in the scalability to
process natural languages. Compositional approaches that are scablable to
natural languages fall short …

arxiv natural networks neural networks