@inproceedings{6f8fa28a45094757a5b5ece84da607f5,
title = "Efficient learning of tier-based strictly k-local languages",
abstract = "We introduce an algorithm that learns the class of Tier-based Strictly k-Local (TSLk) formal languages in polynomial time on a sample of positive data whose size is bounded by a constant. The TSLk languages are useful in modeling the cognition of sound patterns in natural language [6, 11], and it is known that they can be efficiently learned from positive data in the case that k = 2 [9]. We extend this result to any k and improve on its time efficiency. We also refine the definition of a canonical TSLk grammar and prove several properties about these grammars that aid in their learning.",
keywords = "Algorithmic learning, Grammatical inference",
author = "Adam Jardine and Kevin McMullin",
note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 11th International Conference on Language and Automata Theory and Applications, LATA 2017 ; Conference date: 06-03-2017 Through 09-03-2017",
year = "2017",
doi = "10.1007/978-3-319-53733-7_4",
language = "English (US)",
isbn = "9783319537320",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "64--76",
editor = "Frank Drewes and Carlos Mart{\'i}n-Vide and Bianca Truthe",
booktitle = "Language and Automata Theory and Applications - 11th International Conference, LATA 2017, Proceedings",
address = "Germany",
}