Please use this identifier to cite or link to this item: https://doi.org/10.1007/978-3-642-39053-1_21
Title: On conservative learning of recursively enumerable languages
Authors: Gao, Z.
Jain, S. 
Stephan, F. 
Issue Date: 2013
Citation: Gao, Z.,Jain, S.,Stephan, F. (2013). On conservative learning of recursively enumerable languages. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 7921 LNCS : 181-190. ScholarBank@NUS Repository. https://doi.org/10.1007/978-3-642-39053-1_21
Abstract: Conservative partial learning is a variant of partial learning whereby the learner, on a text for a target language L, outputs one index e with L = W e infinitely often and every further hypothesis d is output only finitely often and satisfies L ⊈ Wd. The present paper studies the learning strength of this notion, comparing it with other learnability criteria such as confident partial learning, explanatory learning, as well as behaviourally correct learning. It is further established that for classes comprising infinite sets, conservative partial learnability is in fact equivalent to explanatory learnability relative to the halting problem. © 2013 Springer-Verlag.
Source Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
URI: http://scholarbank.nus.edu.sg/handle/10635/113938
ISBN: 9783642390524
ISSN: 03029743
DOI: 10.1007/978-3-642-39053-1_21
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

7
checked on Apr 2, 2020

Page view(s)

113
checked on Mar 28, 2020

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.