In this paper, we study exact learning of logic programs from entailment queries and present a polynomial time algorithm to learn a rich class of logic programs that allow local variables and include many standard programs like addition, multiplication, exponentiation, member, prefix, suffix, length, append, merge, split, delete, insert, insertionsort, quick-sort, merge-sort, preorder and inorder traversal of binary trees, polynomial recognition, derivatives, sum of a list of naturals. Our algorithm asks at most polynomial number of queries and our class is the largest of all the known classes of programs learnable from entailment. © Springer-Verlag Berlin Heidelberg 2004.
CITATION STYLE
Krishna Rao, M. R. K. (2004). Learnability of simply-moded logic programs from entailment. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3321, 128–141. https://doi.org/10.1007/978-3-540-30502-6_9
Mendeley helps you to discover research relevant for your work.