TY - JOUR
T1 - Learning minimal automata with recurrent neural networks
AU - Aichernig, Bernhard K.
AU - König, Sandra
AU - Mateis, Cristinel
AU - Pferscher, Andrea
AU - Tappler, Martin
N1 - Publisher Copyright:
© The Author(s) 2024.
PY - 2024
Y1 - 2024
N2 - In this article, we present a novel approach to learning finite automata with the help of recurrent neural networks. Our goal is not only to train a neural network that predicts the observable behavior of an automaton but also to learn its structure, including the set of states and transitions. In contrast to previous work, we constrain the training with a specific regularization term. We iteratively adapt the architecture to learn the minimal automaton, in the case where the number of states is unknown. We evaluate our approach with standard examples from the automata learning literature, but also include a case study of learning the finite-state models of real Bluetooth Low Energy protocol implementations. The results show that we can find an appropriate architecture to learn the correct minimal automata in all considered cases.
AB - In this article, we present a novel approach to learning finite automata with the help of recurrent neural networks. Our goal is not only to train a neural network that predicts the observable behavior of an automaton but also to learn its structure, including the set of states and transitions. In contrast to previous work, we constrain the training with a specific regularization term. We iteratively adapt the architecture to learn the minimal automaton, in the case where the number of states is unknown. We evaluate our approach with standard examples from the automata learning literature, but also include a case study of learning the finite-state models of real Bluetooth Low Energy protocol implementations. The results show that we can find an appropriate architecture to learn the correct minimal automata in all considered cases.
KW - Automata learning
KW - Bluetooth Low Energy
KW - Machine learning
KW - Model inference
KW - Recurrent neural networks
UR - http://www.scopus.com/inward/record.url?scp=85188437819&partnerID=8YFLogxK
U2 - 10.1007/s10270-024-01160-6
DO - 10.1007/s10270-024-01160-6
M3 - Article
AN - SCOPUS:85188437819
SN - 1619-1366
JO - Software and Systems Modeling
JF - Software and Systems Modeling
ER -