東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Learning via queries.

Chen, Jiang.

Linked to FindBook

Google Book

Amazon

博客來

Learning via queries.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Learning via queries./
Author:	Chen, Jiang.
Description:	117 p.
Notes:	Source: Dissertation Abstracts International, Volume: 67-04, Section: B, page: 2073.
Contained By:	Dissertation Abstracts International67-04B.
Subject:	Computer Science. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3214192
ISBN:	9780542651403

Learning via queries.
Chen, Jiang.

Learning via queries. - 117 p.

Source: Dissertation Abstracts International, Volume: 67-04, Section: B, page: 2073.

Thesis (Ph.D.)--Yale University, 2006.

Query learning is one of the most important active learning models that have been studied in the literature. In this thesis, we study two types of queries, namely, edge-detecting queries and value-injecting queries, both of which are motivated by biological applications.

ISBN: 9780542651403Subjects--Topical Terms:

626642
Computer Science.

Learning via queries.
LDR:02012nmm 2200289 4500 001 1829014
005 20071106080103.5
008 130610s2006 eng d
020 $a 9780542651403
035 $a (UMI)AAI3214192
035 $a AAI3214192
040 $a UMI $c UMI
100 1 $a Chen, Jiang. $3 1917886
245 1 0 $a Learning via queries.
300 $a 117 p.
500 $a Source: Dissertation Abstracts International, Volume: 67-04, Section: B, page: 2073.
500 $a Adviser: Dana Angluin.
502 $a Thesis (Ph.D.)--Yale University, 2006.
520 $a Query learning is one of the most important active learning models that have been studied in the literature. In this thesis, we study two types of queries, namely, edge-detecting queries and value-injecting queries, both of which are motivated by biological applications.
520 $a Motivated by PCR experiments used in a genome sequencing problem, edge-detecting queries allow the learner to query whether a set of vertices induces an edge or not. We present algorithms that learn general graphs and uniform hypergraphs with queries linear in the number of edges. The algorithms are optimal in terms of their dependence on the number of edges. In both problems, we present algorithms in which queries can be made in a small number of rounds. We also give matching upper and lower bounds for the problem of learning almost uniform hypergraphs.
520 $a Value-injection queries are motivated by experiments that use selective gene disruption and expression to identify gene regulatory networks. We propose a new model of exact learning of acyclic circuits with value-injection queries. We prove lower bounds and show that a large class of circuits including NC1 and AC0 are learnable in the new model.
590 $a School code: 0265.
650 4 $a Computer Science. $3 626642
690 $a 0984
710 2 0 $a Yale University. $3 515640
773 0 $t Dissertation Abstracts International $g 67-04B.
790 1 0 $a Angluin, Dana, $e advisor
790 $a 0265
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3214192