Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples
Machine Learning
2020-02-28 v4 Formal Languages and Automata Theory
Abstract
We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L* algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.
Keywords
Cite
@article{arxiv.1711.09576,
title = {Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples},
author = {Gail Weiss and Yoav Goldberg and Eran Yahav},
journal= {arXiv preprint arXiv:1711.09576},
year = {2020}
}
Comments
Accepted in ICML 2018, (Feb 2020: added link to code, at https://github.com/tech-srl/lstar_extraction )