Abstract
dc:description.abstractLearning Disjunctive Normal Form and Threshold of Parity functions are well-studied problems in computational learning theory. Under different learning models we can achieve different time complexities with respect to the size of the input. In this thesis, I consider using an oracle whose power lies in between membership and example oracles, the random walk oracle. I first study a heuristic algorithm for estimating the sum of squares of Fourier coefficients. Then I propose a new algorithm for learning under a random walk oracle and compare the performance of this algorithm with an algorithm previously proposed by Jackson and Wimmer. The performance of these two algorithms is evaluated empirically based on experiments on linear threshold functions, a special case of TOP. Finally, I prove a theorem showing how to efficiently generate a richer subclass of TOP functions from linear threshold functions, which could allow for broader testing of these algorithms.
Degree
thesis:*- Name thesis:degree_name
- MS
- Level thesis:degree_level
- Campus Only
- Discipline thesis:degree_discipline
- Computational Mathematics
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Yang, Xiaolin
- Contributors dc:contributor
-
- Jeffrey C. Jackson
- Donald L. Simon
Subjects
dc:subject × 3Rights
- Language dc:language
- English
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://dsc.duq.edu/etd/1569