Learning Deterministic Variable Automata over Infinite Alphabets

Sheinvald, Sarai

doi:10.1007/978-3-030-30942-8_37

Cited by 4 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many works study the learnability of symbolic automata, e.g. [5,15,28,43]. The work of [23] studies alphabet refinement for learning symbolic automata.…”

Section: Related Workmentioning

confidence: 99%

Assume, Guarantee or Repair -- A Regular Framework for Non Regular Properties (full version)

Frenkel¹,

Grumberg²,

Păsăreanu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We present Assume-Guarantee-Repair (AGR) -a novel framework which verifies that a program satisfies a set of properties and also repairs the program in case the verification fails. We consider communicating programs -these are simple C-like programs, extended with synchronous actions over communication channels. Our method, which consists of a learning-based approach to assume-guarantee reasoning, performs verification and repair simultaneously: in every iteration, AGR either makes another step towards proving that the (current) system satisfies the required properties, or alters the system in a way that brings it closer to satisfying the properties. To handle infinite-state systems we build finite abstractions, for which we check the satisfaction of complex properties that contain first-order constraints, using both syntactic and semantic-aware methods. We implemented AGR and evaluated it on various communication protocols. Our experiments present compact proofs of correctness and quick repairs.

show abstract

“…Many works study the learnability of symbolic automata, e.g. [5,15,28,43]. The work of [23] studies alphabet refinement for learning symbolic automata.…”

Section: Related Workmentioning

confidence: 99%

Assume, Guarantee or Repair -- A Regular Framework for Non Regular Properties (full version)

Frenkel¹,

Grumberg²,

Păsăreanu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent years, the software engineering research community has used grammatical inference [21,25,28,29] because of its possibility to solve a wide range of practical applications of formal verication [12], model inference [11,20,22,24,30] and software testing [4]. These applications in general use the concept of inferring an automaton by generating a model of a system under learn (SUL) and analyzing it to check its behavioral correctness with respect to a specification.…”

Section: Introductionmentioning

confidence: 99%

DLIQ: A Deterministic Finite Automaton Learning Algorithm through Inverse Queries

Haneef

Sindhu

2022

ITC

View full text Add to dashboard Cite

Automaton learning has attained a renewed interest in many interesting areas of software engineering including formal verification, software testing and model inference. An automaton learning algorithm typically learns the regular language of a DFA with the help of queries. These queries are posed by the learner (Learning Algorithm) to a Minimally Adequate Teacher (MAT). The MAT can generally answer two types of queries asked by the learning algorithm; membership queries and equivalence queries. Learning algorithms can be categorized into two broad categories: incremental and complete learning algorithms. Likewise, these can be designed for 1-bit learning or k-bit learning. Existing automaton learning algorithms have polynomial (atleast cubic) time complexity in the presence of a MAT. Therefore, sometimes these algorithms even become fail to learn large complex software systems. In this research work, we have reduced the complexity of the Deterministic Finite Automaton (DFA) learning into lower bounds (from cubic to square form). For this, we introduce an efficient complete DFA learning algorithm through Inverse Queries (DLIQ) based on the concept of inverse queries introduced by John Hopcroft for state minimization of a DFA. The DLIQ algorithm takes O(|Ps||F|+|Σ|N) complexity in the presence of a MAT which is also equipped to answer inverse queries. We give a theoretical analysis of the proposed algorithm along with providing a proof correctness and termination of the DLIQ algorithm. We also compare the performance of DLIQ with ID algorithm by implementing an evaluation framework. Our results depict that DLIQ is more efficient than ID algorithm in terms of time complexity.

show abstract

“…There already exists substantial literature on learning restricted forms of SFAs [GJL10, MM14, ASKK16, MM17, CDYS17], as well as general SFAs [DD17,AD18], and even nondeterministic residual SFAs [CHYS19]. For other types of automata over infinite alphabets, [HSM11] suggests learning abstractions, and [She19] presents a learning algorithm for deterministic variable automata. All these works consider the query learning paradigm, and provide extensions to Angluin's L * algorithm for learning DFAs using membership and equivalence queries [Ang87].…”

Section: Introductionmentioning

confidence: 99%

Inferring Symbolic Automata

Fisman¹,

Frenkel²,

Zilles³

2021

Preprint

View full text Add to dashboard Cite

We study the learnability of symbolic finite state automata (SFA), a model shown useful in many applications in software verification. The state-of-the-art literature on this topic follows the query learning paradigm, and so far all obtained results are positive. We provide a necessary condition for efficient learnability of SFAs in this paradigm, from which we obtain the first negative result.The main focus of our work lies in the learnability of SFAs under the paradigm of identification in the limit using polynomial time and data, and its strengthening efficiently identifiability, which are concerned with the existence of a systematic set of characteristic samples from which a learner can correctly infer the target language. We provide a necessary condition for identification of SFAs in the limit using polynomial time and data, and a sufficient condition for efficient learnability of SFAs. From these conditions we derive a positive and a negative result.Since the performance of a learning algorithm is bounded as a function of the size of the representation of the target language and SFAs, in general, do not have a canonical form, and there are trade-offs between the complexity of the predicates on the transitions and the number of transitions, we start by defining size measures for SFAs. We revisit the complexity of procedures on SFAs and analyze them according to these measures, paying attention to the special forms of SFAs: normalized SFAs and neat SFAs, as well as to SFAs over a monotonic effective Boolean algebra.

show abstract

Learning Deterministic Variable Automata over Infinite Alphabets

Cited by 4 publications

References 11 publications

Assume, Guarantee or Repair -- A Regular Framework for Non Regular Properties (full version)

Assume, Guarantee or Repair -- A Regular Framework for Non Regular Properties (full version)

DLIQ: A Deterministic Finite Automaton Learning Algorithm through Inverse Queries

Inferring Symbolic Automata

Contact Info

Product

Resources

About