Learning Nondeterministic Real-Time Automata

被引:0
作者
An, Jie [1 ]
Zhan, Bohua [2 ,3 ,4 ]
Zhan, Naijun [2 ,3 ,4 ]
Zhang, Miaomiao [5 ]
机构
[1] Max Planck Inst Software Syst, Paul Ehrlich Str G 26, D-67663 Kaiserslautern, Germany
[2] Chinese Acad Sci, SKLCS, Inst Software, Beijing 100190, Peoples R China
[3] Chinese Acad Sci, Sci & Technol Integrated Informat Syst Lab, Inst Software, Beijing 100190, Peoples R China
[4] Univ Chinese Acad Sci, Beijing 100049, Peoples R China
[5] Tongji Univ, Sch Software Engn, Shanghai 100049, Peoples R China
关键词
Active learning; model learning; nondeterministic real-time automata; real-time languages; CHECKING;
D O I
10.1145/3477030
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
We present an active learning algorithm named NRTALearning for nondeterministic real-time automata (NRTAs). Real-time automata (RTAs) are a subclass of timed automata with only one clock which resets at each transition. First, we prove the corresponding Myhill-Nerode theorem for real-time languages. Then we show that there exists a unique minimal deterministic real-time automaton (DRTA) recognizing a given real-time language, but the same does not hold for NRTAs. We thus define a special kind of NRTAs, named residual real-time automata (RRTAs), and prove that there exists a minimal RRTA to recognize any given real-time language. This transforms the learning problem of NRTAs to the learning problem of RRTAs. After describing the learning algorithm in detail, we prove its correctness and polynomial complexity. In addition, based on the corresponding Myhill-Nerode theorem, we extend the existing active learning algorithm NL* for nondeterministic finite automata to learn RRTAs. We evaluate and compare the two algorithms on two benchmarks consisting of randomly generated NRTAs and rational regular expressions. The results show that NRTALearning generally performs fewer membership queries and more equivalence queries than the extended NL* algorithm, and the learnt NRTAs have much fewer locations than the corresponding minimal DRTAs. We also conduct a case study using a model of scheduling of final testing of integrated circuits.
引用
收藏
页数:26
相关论文
共 35 条
  • [1] From Passive to Active: Learning Timed Automata Efficiently
    Aichernig, Bernhard K.
    Pferscher, Andrea
    Tappler, Martin
    [J]. NASA FORMAL METHODS (NFM 2020), 2020, 12229 : 3 - 21
  • [2] A THEORY OF TIMED AUTOMATA
    ALUR, R
    DILL, DL
    [J]. THEORETICAL COMPUTER SCIENCE, 1994, 126 (02) : 183 - 235
  • [3] Alur R, 2005, LECT NOTES COMPUT SC, V3576, P548
  • [4] Learning real-time automata
    An, Jie
    Wang, Lingtai
    Zhan, Bohua
    Zhan, Naijun
    Zhang, Miaomiao
    [J]. SCIENCE CHINA-INFORMATION SCIENCES, 2021, 64 (09)
  • [5] An J, 2020, LECT NOTES COMPUT SC, V12078, P444, DOI 10.1007/978-3-030-45190-5_25
  • [6] Model Checking Bounded Continuous-time Extended Linear Duration Invariants
    An, Jie
    Zhan, Naijun
    Li, Xiaoshan
    Zhang, Miaomiao
    Yi, Wang
    [J]. HSCC 2018: PROCEEDINGS OF THE 21ST INTERNATIONAL CONFERENCE ON HYBRID SYSTEMS: COMPUTATION AND CONTROL (PART OF CPS WEEK), 2018, : 81 - 90
  • [7] LEARNING REGULAR SETS FROM QUERIES AND COUNTEREXAMPLES
    ANGLUIN, D
    [J]. INFORMATION AND COMPUTATION, 1987, 75 (02) : 87 - 106
  • [8] [Anonymous], 2010, THESIS DELFT U TECHN
  • [9] Bollig B, 2009, 21ST INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI-09), PROCEEDINGS, P1004
  • [10] Checking NFA Equivalence with Bisimulations up to Congruence
    Bonchi, Filippo
    Pous, Damien
    [J]. ACM SIGPLAN NOTICES, 2013, 48 (01) : 457 - 468