LOMay 14, 2024
Output-decomposed Learning of Mealy MachinesRick Koenders, Joshua Moerman
We present an active automata learning algorithm which learns a decomposition of a finite state machine, based on projecting onto individual outputs. This is dual to a recent compositional learning algorithm by Labbaf et al. (2023). When projecting the outputs to a smaller set, the model itself is reduced in size. By having several such projections, we do not lose any information and the full system can be reconstructed. Depending on the structure of the system this reduces the number of queries drastically, as shown by a preliminary evaluation of the algorithm.
SEMay 8, 2017
Learning Product AutomataJoshua Moerman
In this paper we give an optimization for active learning algorithms, applicable to learning Moore machines where the output comprises several observables. These machines can be decomposed themselves by projecting on each observable, resulting in smaller components. These components can then be learnt with fewer queries. This is in particular interesting for learning software, where compositional methods are important for guaranteeing scalability.
SENov 8, 2016
Complementing Model Learning with Mutation-Based FuzzingRick Smetsers, Joshua Moerman, Mark Janssen et al.
An ongoing challenge for learning algorithms formulated in the Minimally Adequate Teacher framework is to efficiently obtain counterexamples. In this paper we compare and combine conformance testing and mutation-based fuzzing methods for obtaining counterexamples when learning finite state machine models for the reactive software systems of the Rigorous Exampination of Reactive Systems (RERS) challenge. We have found that for the LTL problems of the challenge the fuzzer provided an independent confirmation that the learning process had been successful, since no additional counterexamples were found. For the reachability problems of the challenge, however, the fuzzer discovered more reachable error states than the learner and tester, albeit in some cases the learner and tester found some that were not discovered by the fuzzer. This leads us to believe that these orthogonal approaches are complementary in the context of model learning.