Projects per year
Abstract
In this work, we study L∗-based learning of deterministic Markov decision processes, a class of Markov decision processes where an observation following an action uniquely determines a successor state. For this purpose, we first assume an ideal setting with a teacher who provides perfect information to the student. Then, we relax this assumption and present a novel learning algorithm that collects information by sampling execution traces of the system via testing.
Experiments performed on an implementation of our sampling-based algorithm suggest that our method achieves better accuracy than state-of-the-art passive learning techniques using the same amount of test obser vations. In contrast to existing learning algorithms which assume a predefined number of states, our algorithm learns the complete model structure including the state space.
Original language | English |
---|---|
Pages (from-to) | 575-615 |
Number of pages | 41 |
Journal | Formal Aspects of Computing |
Volume | 33 |
Issue number | 4-5 |
Early online date | 31 Mar 2021 |
DOIs | |
Publication status | Published - Aug 2021 |
Keywords
- Active automata learning
- Markov decision processes
- Model inference
ASJC Scopus subject areas
- Software
- Theoretical Computer Science
Fields of Expertise
- Information, Communication & Computing
Fingerprint
Dive into the research topics of 'L*-Based Learning of Markov Decision Processes (Extended Version)'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Dependable Internet of Things
Boano, C. A. (Co-Investigator (CoI)), Kubin, G. (Co-Investigator (CoI)), Bloem, R. (Co-Investigator (CoI)), Horn, M. (Co-Investigator (CoI)), Pernkopf, F. (Co-Investigator (CoI)), Zakany, N. (Co-Investigator (CoI)), Mangard, S. (Co-Investigator (CoI)), Witrisal, K. (Co-Investigator (CoI)), Römer, K. U. (Co-Investigator (CoI)), Aichernig, B. (Co-Investigator (CoI)), Bösch, W. (Co-Investigator (CoI)), Baunach, M. C. (Co-Investigator (CoI)), Tappler, M. (Co-Investigator (CoI)), Malenko, M. (Co-Investigator (CoI)), Weiser, S. (Co-Investigator (CoI)), Eichlseder, M. (Co-Investigator (CoI)), Leitinger, E. (Co-Investigator (CoI)), Grosinger, J. (Co-Investigator (CoI)), Großwindhager, B. (Co-Investigator (CoI)), Ebrahimi, M. (Co-Investigator (CoI)), Alothman Alterkawi, A. B. (Co-Investigator (CoI)), Knoll, C. (Co-Investigator (CoI)), Teschl, R. (Co-Investigator (CoI)), Saukh, O. (Co-Investigator (CoI)), Rath, M. (Co-Investigator (CoI)), Steinberger, M. (Co-Investigator (CoI)), Steinbauer-Wagner, G. (Co-Investigator (CoI)) & Tranninger, M. (Co-Investigator (CoI))
1/01/16 → 31/03/22
Project: Research project
Research output
- 1 Conference paper
-
L*-Based Learning of Markov Decision Processes
Tappler, M., Aichernig, B., Bacci, G., Eichlseder, M. & Larsen, K. G., 2019, Formal Methods - The Next 30 Years . ter Beek, M. H., McIver, A. & Oliveria, J. N. (eds.). Cham: Springer, p. 651 - 669 19 p. (Lecture Notes in Computer Science; vol. 11800).Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review
Open Access