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 structured systems from imp...

Goldowsky, Boris Neill.

Linked to FindBook

Google Book

Amazon

博客來

Learning structured systems from imperfect information.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Learning structured systems from imperfect information./
Author:	Goldowsky, Boris Neill.
Description:	64 p.
Notes:	Source: Dissertation Abstracts International, Volume: 56-05, Section: B, page: 2910.
Contained By:	Dissertation Abstracts International56-05B.
Subject:	Language, Linguistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9530388

Learning structured systems from imperfect information.
Goldowsky, Boris Neill.

Learning structured systems from imperfect information. - 64 p.

Source: Dissertation Abstracts International, Volume: 56-05, Section: B, page: 2910.

Thesis (Ph.D.)--University of Rochester, 1995.

A critical way in which language acquisition has been said to differ from other learning is in the results of inconsistent input. In language acquisition, it is common for the data available to the learner to be imperfect, but this has little effect on children's learning--they force the system to be regular by extracting or creating rules. In other domains, however, such as probability-learning experiments, people are said to be faithful to the probabilistic structure of the data to which they are exposed. This leads to the assumption that language must be learned by different mechanisms.Subjects--Topical Terms:

1018079
Language, Linguistics.

Learning structured systems from imperfect information.
LDR:02751nam 2200313 a 45 001 970147
005 20110921
008 110921s1995 eng d
035 $a (UMI)AAI9530388
035 $a AAI9530388
040 $a UMI $c UMI
100 1 $a Goldowsky, Boris Neill. $3 1294194
245 1 0 $a Learning structured systems from imperfect information.
300 $a 64 p.
500 $a Source: Dissertation Abstracts International, Volume: 56-05, Section: B, page: 2910.
500 $a Supervisor: Elissa Newport.
502 $a Thesis (Ph.D.)--University of Rochester, 1995.
520 $a A critical way in which language acquisition has been said to differ from other learning is in the results of inconsistent input. In language acquisition, it is common for the data available to the learner to be imperfect, but this has little effect on children's learning--they force the system to be regular by extracting or creating rules. In other domains, however, such as probability-learning experiments, people are said to be faithful to the probabilistic structure of the data to which they are exposed. This leads to the assumption that language must be learned by different mechanisms.
520 $a However, previous studies may have confounded the effect of linguistic vs. non-linguistic tasks with the complexity or structure of the learning problem, or with the age of the learners studied. Following Newport's Less is More hypothesis, we suggest that the structure of the system being learned interacts with the capabilities and constraints of the learner to create the different learning patterns, independent of the domain.
520 $a This dissertation describes three experiments, which use a new paradigm for presenting multidimensional structured systems. The subjects are to learn a mapping relation between objects and actions by watching observation trials which exemplify the regularities but also contain exceptions. Adults and 7-year-old children are tested on various systems which vary in complexity and quality of data, and their learning is evaluated by prediction and generalization tests.
520 $a While adults turn out to be better at learning the details of the systems, children demonstrate a tendency to respond according to a consistent pattern, inventing one if necessary. This is parallel to some phenomena of language acquisition, suggesting that language may be learned by the same mechanisms that support the learning of other complex systems.
590 $a School code: 0188.
650 4 $a Language, Linguistics. $3 1018079
650 4 $a Psychology, Developmental. $3 1017557
650 4 $a Psychology, Experimental. $3 517106
690 $a 0290
690 $a 0620
690 $a 0623
710 2 0 $a University of Rochester. $3 515736
773 0 $t Dissertation Abstracts International $g 56-05B.
790 $a 0188
790 1 0 $a Newport, Elissa, $e advisor
791 $a Ph.D.
792 $a 1995
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9530388