Communications of the ACM
On learning from queries and counterexamples in the presence of noise
Information Processing Letters
Learning in the presence of finitely or infinitely many irrelevant attributes
COLT '91 Proceedings of the fourth annual workshop on Computational learning theory
Learning read-once formulas with queries
Journal of the ACM (JACM)
Learning with malicious membership queries and exceptions (extended abstract)
COLT '94 Proceedings of the seventh annual conference on Computational learning theory
Learning with queries but incomplete information (extended abstract)
COLT '94 Proceedings of the seventh annual conference on Computational learning theory
Exact learning Boolean functions via the monotone theory
Information and Computation
Simple learning algorithms using divide and conquer
COLT '95 Proceedings of the eighth annual conference on Computational learning theory
Malicious Omissions and Errors in Answers to Membership Queries
Machine Learning
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
Attribute-efficient learning in query and mistake-bound models
Journal of Computer and System Sciences
On-line learning with malicious noise and the closure algorithm
Annals of Mathematics and Artificial Intelligence
Machine Learning
Machine Learning
Machine Learning
Learning with Errors in Answers to Membership Queries (Extracted Abstract)
FOCS '04 Proceedings of the 45th Annual IEEE Symposium on Foundations of Computer Science
Learning with errors in answers to membership queries
Journal of Computer and System Sciences
| Hi-index | 0.00 |
We study learning in a modified EXACT model, where the oracles are corrupt and only few of the presented attributes are relevant. Both modifications were already studied in the literature, and efficient solutions were found to most of their variants. Nonetheless, their reasonable combination is yet to be studied, and combining the existing solutions either fails or works with complexity that can be significantly improved. In this paper we prove equivalence of EXACT learning attribute-efficiently with and without corrupt oracles. For each of the possible scenarios we describe a generic scheme that enables learning in these cases using modifications of the standard learning algorithms. We also generalize and improve previous non attribute-efficient algorithms for learning with corrupt oracles.