Query Learning Automata with Helpful Labels

Adrian-Horia Dediu, Joana M. Matos, Claudio Moraga
; Proceedings of The 13th International Conference on Grammatical Inference, PMLR 57:15-29, 2017.

Abstract

In the active learning framework, a modified query learning algorithm benefiting by a nontrivial helpful labeling is able to learn automata with a reduced number of queries. In extremis, there exists a helpful labeling allowing the algorithm to learn automata even without counterexamples. We also review the correction queries defining them as particular types of labeling. We introduce minimal corrections, maximal corrections, and random corrections. An experimental approach compares the performance and limitations of various types of queries and corrections. The results show that algorithms using corrections require fewer queries in most of the cases.

Cite this Paper


BibTeX
@InProceedings{pmlr-v57-dediu16, title = {Query Learning Automata with Helpful Labels}, author = {Adrian-Horia Dediu and Joana M. Matos and Claudio Moraga}, booktitle = {Proceedings of The 13th International Conference on Grammatical Inference}, pages = {15--29}, year = {2017}, editor = {Sicco Verwer and Menno van Zaanen and Rick Smetsers}, volume = {57}, series = {Proceedings of Machine Learning Research}, address = {Delft, The Netherlands}, month = {05--07 Oct}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v57/dediu16.pdf}, url = {http://proceedings.mlr.press/v57/dediu16.html}, abstract = {In the active learning framework, a modified query learning algorithm benefiting by a nontrivial helpful labeling is able to learn automata with a reduced number of queries. In extremis, there exists a helpful labeling allowing the algorithm to learn automata even without counterexamples. We also review the correction queries defining them as particular types of labeling. We introduce minimal corrections, maximal corrections, and random corrections. An experimental approach compares the performance and limitations of various types of queries and corrections. The results show that algorithms using corrections require fewer queries in most of the cases.} }
Endnote
%0 Conference Paper %T Query Learning Automata with Helpful Labels %A Adrian-Horia Dediu %A Joana M. Matos %A Claudio Moraga %B Proceedings of The 13th International Conference on Grammatical Inference %C Proceedings of Machine Learning Research %D 2017 %E Sicco Verwer %E Menno van Zaanen %E Rick Smetsers %F pmlr-v57-dediu16 %I PMLR %J Proceedings of Machine Learning Research %P 15--29 %U http://proceedings.mlr.press %V 57 %W PMLR %X In the active learning framework, a modified query learning algorithm benefiting by a nontrivial helpful labeling is able to learn automata with a reduced number of queries. In extremis, there exists a helpful labeling allowing the algorithm to learn automata even without counterexamples. We also review the correction queries defining them as particular types of labeling. We introduce minimal corrections, maximal corrections, and random corrections. An experimental approach compares the performance and limitations of various types of queries and corrections. The results show that algorithms using corrections require fewer queries in most of the cases.
RIS
TY - CPAPER TI - Query Learning Automata with Helpful Labels AU - Adrian-Horia Dediu AU - Joana M. Matos AU - Claudio Moraga BT - Proceedings of The 13th International Conference on Grammatical Inference PY - 2017/01/16 DA - 2017/01/16 ED - Sicco Verwer ED - Menno van Zaanen ED - Rick Smetsers ID - pmlr-v57-dediu16 PB - PMLR SP - 15 DP - PMLR EP - 29 L1 - http://proceedings.mlr.press/v57/dediu16.pdf UR - http://proceedings.mlr.press/v57/dediu16.html AB - In the active learning framework, a modified query learning algorithm benefiting by a nontrivial helpful labeling is able to learn automata with a reduced number of queries. In extremis, there exists a helpful labeling allowing the algorithm to learn automata even without counterexamples. We also review the correction queries defining them as particular types of labeling. We introduce minimal corrections, maximal corrections, and random corrections. An experimental approach compares the performance and limitations of various types of queries and corrections. The results show that algorithms using corrections require fewer queries in most of the cases. ER -
APA
Dediu, A., M. Matos, J. & Moraga, C.. (2017). Query Learning Automata with Helpful Labels. Proceedings of The 13th International Conference on Grammatical Inference, in PMLR 57:15-29

Related Material