Abstract
In this paper, 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 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 automata in all considered cases.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The actual corresponding input, resp. output, symbol is obtained from the input, resp. output, symbol alphabet through an appropriate indexed mapping. For simplicity, we don’t show this mapping here.
References
Aichernig, B.K., Mostowski, W., Mousavi, M.R., Tappler, M., Taromirad, M.: Model learning and model-based testing. In: Bennaceur, A., Hähnle, R., Meinke, K. (eds.) Machine Learning for Dynamic Software Analysis: Potentials and Limits. LNCS, vol. 11026, pp. 74–100. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-96562-8_3
Angluin, D.: Learning regular sets from queries and counterexamples. Inf. Comput. 75(2), 87–106 (1987). https://doi.org/10.1016/0890-5401(87)90052-6
Carr, S., Jansen, N., Topcu, U.: Verifiable RNN-based policies for POMDPs under temporal logic constraints. In: IJCAI, pp. 4121–4127. ijcai.org (2020). https://doi.org/10.24963/ijcai.2020/570
Dong, G., et al.: Towards interpreting recurrent neural networks through probabilistic abstraction. In: ASE, pp. 499–510. IEEE (2020). https://doi.org/10.1145/3324884.3416592
Elman, J.L.: Finding structure in time. Cogn. Sci. 14(2), 179–211 (1990). https://doi.org/10.1207/s15516709cog1402_1
Gold, E.M.: Complexity of automaton identification from given data. Inf. Control 37(3), 302–320 (1978). https://doi.org/10.1016/S0019-9958(78)90562-4
Goudreau, M.W., Giles, C.L., Chakradhar, S.T., Chen, D.: First-order versus second-order single-layer recurrent neural networks. IEEE Trans. Neural Netw. 5(3), 511–513 (1994). https://doi.org/10.1109/72.286928
Heule, M., Verwer, S.: Software model synthesis using satisfiability solvers. Empir. Softw. Eng. 18(4), 825–856 (2013)
de la Higuera, C.: Grammatical Inference: Learning Automata and Grammars. Cambridge University Press, New York (2010)
Howar, F., Steffen, B.: Active automata learning in practice. In: Bennaceur, A., Hähnle, R., Meinke, K. (eds.) Machine Learning for Dynamic Software Analysis: Potentials and Limits. LNCS, vol. 11026, pp. 123–148. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-96562-8_5
Irfan, M.N., Oriat, C., Groz, R.: Model inference and testing. In: Advances in Computers, vol. 89, pp. 89–139. Elsevier (2013)
Isberner, M., Howar, F., Steffen, B.: The open-source LearnLib. In: Kroening, D., Păsăreanu, C.S. (eds.) CAV 2015, Part I. LNCS, vol. 9206, pp. 487–495. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-21690-4_32
Khmelnitsky, I., et al.: Property-directed verification and robustness certification of recurrent neural networks. In: Hou, Z., Ganesh, V. (eds.) ATVA 2021. LNCS, vol. 12971, pp. 364–380. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-88885-5_24
Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. In: ICLR (2015)
Kleene, S.C.: Representation of Events in Nerve Nets and Finite Automata. RAND Corporation, Santa Monica (1951)
Koul, A., Fern, A., Greydanus, S.: Learning finite state representations of recurrent policy networks. In: ICLR. OpenReview.net (2019)
Ma, Y., Principe, J.C.: A taxonomy for neural memory networks. IEEE Trans. Neural Netw. Learn. Syst. 31(6), 1780–1793 (2020). https://doi.org/10.1109/TNNLS.2019.2926466
Mayr, F., Yovine, S.: Regular inference on artificial neural networks. In: Holzinger, A., Kieseberg, P., Tjoa, A.M., Weippl, E. (eds.) CD-MAKE 2018. LNCS, vol. 11015, pp. 350–369. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-99740-7_25
Michalenko, J.J., Shah, A., Verma, A., Baraniuk, R.G., Chaudhuri, S., Patel, A.B.: Representing formal languages: A comparison between finite automata and recurrent neural networks. In: ICLR. OpenReview.net (2019)
Minsky, M.L.: Computation: Finite and Infinite Machines. Prentice-Hall Inc., USA (1967)
Muškardin, E., Aichernig, B.K., Pill, I., Pferscher, A., Tappler, M.: AALpy: An active automata learning library. In: Hou, Z., Ganesh, V. (eds.) ATVA 2021. LNCS, vol. 12971, pp. 67–73. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-88885-5_5
Muskardin, E., Aichernig, B.K., Pill, I., Tappler, M.: Learning finite state models from recurrent neural networks. In: ter Beek, M.H., Monahan, R. (eds.) IFM 2022. LNCS, vol. 13274, pp. 229–248. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-07727-2_13
Oliva, C., Lago-Fernández, L.F.: Stability of internal states in recurrent neural networks trained on regular languages. Neurocomputing 452, 212–223 (2021). https://doi.org/10.1016/j.neucom.2021.04.058
Omlin, C.W., Giles, C.L.: Extraction of rules from discrete-time recurrent neural networks. Neural Netw. 9(1), 41–52 (1996). https://doi.org/10.1016/0893-6080(95)00086-0
Oncina, J., Garcia, P.: Identifying regular languages in polynomial time. In: Advances in Structural and Syntactic Pattern Recognition. Machine Perception and Artificial Intelligence, vol. 5, pp. 99–108. World Scientific (1992)
Paszke, A., et al.: PyTorch: An imperative style, high-performance deep learning library. In: NeurIPS, pp. 8024–8035. Curran Associates, Inc. (2019)
Pferscher, A., Aichernig, B.K.: Fingerprinting Bluetooth Low Energy devices via active automata learning. In: Huisman, M., Păsăreanu, C., Zhan, N. (eds.) FM 2021. LNCS, vol. 13047, pp. 524–542. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-90870-6_28
Shahbaz, M., Groz, R.: Inferring Mealy machines. In: Cavalcanti, A., Dams, D.R. (eds.) FM 2009. LNCS, vol. 5850, pp. 207–222. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-05089-3_14
Smetsers, R., Fiterău-Broştean, P., Vaandrager, F.: Model learning as a satisfiability modulo theories problem. In: Klein, S.T., Martín-Vide, C., Shapira, D. (eds.) LATA 2018. LNCS, vol. 10792, pp. 182–194. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-77313-1_14
Tappler, M., Aichernig, B.K., Lorber, F.: Timed automata learning via SMT solving. In: Deshmukh, J.V., Havelund, K., Perez, I. (eds.) NFM 2022. LNCS, vol. 13260, pp. 489–507. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-06773-0_26
Tin̆o, P., S̆ajda, J.: Learning and extracting initial Mealy automata with a modular neural network model. Neural Comput. 7(4), 822–844 (1995). https://doi.org/10.1162/neco.1995.7.4.822
Tomita, M.: Dynamic construction of finite automata from examples using hill-climbing. In: Conference of the Cognitive Science Society, pp. 105–108 (1982)
Weiss, G., Goldberg, Y., Yahav, E.: Extracting automata from recurrent neural networks using queries and counterexamples. In: ICML. Proceedings of Machine Learning Research, vol. 80, pp. 5244–5253. PMLR (2018)
Weiss, G., Goldberg, Y., Yahav, E.: Learning deterministic weighted automata with queries and counterexamples. In: NeurIPS, pp. 8558–8569 (2019)
Yellin, D.M., Weiss, G.: Synthesizing context-free grammars from recurrent neural networks. In: Groote, J.F., Larsen, K.G. (eds.) TACAS 2021. LNCS, vol. 12651, pp. 351–369. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-72016-2_19
Acknowledgement
This work was collaboratively done in the TU Graz LEAD project Dependable Internet of Things in Adverse Environments project, the LearnTwins project funded by FFG (Österreichische Forschungsförderungsgesellschaft) under grant 880852, and the “University SAL Labs” initiative of Silicon Austria Labs (SAL) and its Austrian partner universities for applied fundamental research for electronic based systems.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Aichernig, B.K., König, S., Mateis, C., Pferscher, A., Schmidt, D., Tappler, M. (2022). Constrained Training of Recurrent Neural Networks for Automata Learning. In: Schlingloff, BH., Chai, M. (eds) Software Engineering and Formal Methods. SEFM 2022. Lecture Notes in Computer Science, vol 13550. Springer, Cham. https://doi.org/10.1007/978-3-031-17108-6_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-17108-6_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-17107-9
Online ISBN: 978-3-031-17108-6
eBook Packages: Computer ScienceComputer Science (R0)