Learning regular sets from queries and counterexamples
Information and Computation
On learning from queries and counterexamples in the presence of noise
Information Processing Letters
Topics in the theory of machine learning and neural computing
Topics in the theory of machine learning and neural computing
Learning k-term DNF formulas with an incomplete membership oracle
COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
On the necessity of Occam algorithms
Theoretical Computer Science
Lower Bound Methods and Separation Results for On-Line Learning Models
Machine Learning - Computational learning theory
Exact identification of read-once formulas using fixed points of amplification functions
SIAM Journal on Computing
Learning decision trees using the Fourier spectrum
SIAM Journal on Computing
Efficient noise-tolerant learning from statistical queries
STOC '93 Proceedings of the twenty-fifth annual ACM symposium on Theory of computing
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
Learning from a consistently ignorant teacher
COLT '94 Proceedings of the seventh annual conference on Computational learning theory
Randomly Fallible Teachers: Learning Monotone DNF with an Incomplete Membership Oracle
Machine Learning - Special issue on computational learning theory
Simulating access to hidden information while learning
STOC '94 Proceedings of the twenty-sixth annual ACM symposium on Theory of computing
Learning Fallible Deterministic Finite Automata
Machine Learning - Special issue on COLT '93
Machine Learning
Machine Learning
Improved lower bounds for learning from noisy examples: an information-theoretic approach
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
The query complexity of finding local minima in the lattice
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
The query complexity of finding local minima in the lattice
Information and Computation
Learning Intermediate Concepts
ALT '01 Proceedings of the 12th International Conference on Algorithmic Learning Theory
A Theoretical Analysis of Query Selection for Collaborative Filtering
COLT '01/EuroCOLT '01 Proceedings of the 14th Annual Conference on Computational Learning Theory and and 5th European Conference on Computational Learning Theory
Learning Regular Sets with an Incomplete Membership Oracle
COLT '01/EuroCOLT '01 Proceedings of the 14th Annual Conference on Computational Learning Theory and and 5th European Conference on Computational Learning Theory
Theoretical Computer Science
Learning attribute-efficiently with corrupt oracles
Theoretical Computer Science
Learning with errors in answers to membership queries
Journal of Computer and System Sciences
Separating Models of Learning with Faulty Teachers
ALT '07 Proceedings of the 18th international conference on Algorithmic Learning Theory
Active learning with multiple views
Journal of Artificial Intelligence Research
Learning attribute-efficiently with corrupt oracles
ALT'05 Proceedings of the 16th international conference on Algorithmic Learning Theory
| Hi-index | 0.00 |
We consider two issues in polynomial-time exact learning of conceptsusing membership and equivalence queries: (1) errors or omissions inanswers to membership queries, and (2) learning finite variants ofconcepts drawn from a learnable class.To study (1), we introduce two new kinds of membership queries:limited membership queries and malicious membership queries. Each isallowed to give incorrect responses on a maliciously chosen set ofstrings in the domain. Instead of answering correctly about astring, a limited membership query may give a special “I don‘tknow” answer, while a malicious membership query may give the wronganswer. A new parameter L is used to bound the length of anencoding of the set of strings that receive such incorrect answers.Equivalence queries are answered correctly, and learning algorithmsare allowed time polynomial in the usual parameters and L. Anyclass of concepts learnable in polynomial time using equivalence andmalicious membership queries is learnable in polynomial time usingequivalence and limited membership queries; the converse is an openproblem. For the classes of monotone monomials and monotone k-term DNF formulas, we present polynomial-time learning algorithms usinglimited membership queries alone. We present polynomial-timelearning algorithms for the class of monotone DNF formulas usingequivalence and limited membership queries, and using equivalence andmalicious membership queries.To study (2), we consider classes of concepts that are polynomiallyclosed under finite exceptions and a natural operation to addexception tables to a class of concepts. Applying this operation, weobtain the class of monotone DNF formulas with finite exceptions. Wegive a polynomial-time algorithm to learn the class of monotone DNFformulas with finite exceptions using equivalence and membershipqueries. We also give a general transformation showing that anyclass of concepts that is polynomially closed under finite exceptionsand is learnable in polynomial time using standard membership andequivalence queries is also polynomial-time learnable using maliciousmembership and equivalence queries. Corollaries include thepolynomial-time learnability of the following classes using maliciousmembership and equivalence queries: deterministic finite acceptors,boolean decision trees, and monotone DNF formulas with finiteexceptions.