Active Automata Learning: From DFAs to Interface Programs and Beyond

11th International Conference on Grammatical Inference

Bernhard Steffen [e-mail] ; Malte Isberner [e-mail]

PID  :  120769

来源: CEUR

【 摘 要 】

This paper reviews the development of active learning in the last decade under the per spective of treating of data, a major source of undecidability, and therefore a key problem to achieve practicality. Starting with the first case studies, in which data was completely disregarded, we revisit different steps towards dealing with data explicitly in active learn ing: We discuss Mealy Machines as a model for systems with (data) output, automated alphabet abstraction refinement as a twodimensional extension of the partitionrefinement based approach of active learning for inferring not only states but also optimal alphabet abstractions, and Register Mealy Machines, which can be regarded as programs restricted to dataindependent data processing as it is typical for protocols or interface programs. We are convinced that this development has the potential to transform active automata learning into a technology of high practical importance.