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

Neural networks in dynamical systems...

Levin, Asriel Uzi.

Linked to FindBook

Google Book

Amazon

博客來

Neural networks in dynamical systems: A system theoretic approach.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Neural networks in dynamical systems: A system theoretic approach./
Author:	Levin, Asriel Uzi.
Description:	150 p.
Notes:	Source: Dissertation Abstracts International, Volume: 54-01, Section: B, page: 0468.
Contained By:	Dissertation Abstracts International54-01B.
Subject:	Artificial Intelligence. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9315226

Neural networks in dynamical systems: A system theoretic approach.
Levin, Asriel Uzi.

Neural networks in dynamical systems: A system theoretic approach. - 150 p.

Source: Dissertation Abstracts International, Volume: 54-01, Section: B, page: 0468.

Thesis (Ph.D.)--Yale University, 1992.

Artificial neural networks is a rapidly developing field of research which concerns the study of massively parallel structures composed of simple non-linear processors that are interconnected via a set of modifiable connections. The motivation behind the study of these structures is based on empirical evidence that has been collected through the years about the organization of biological nervous systems. By building networks in whose structure we incorporate basic features of the biological nervous system, we hope to be able to synthesize artificial systems that exhibit some of the brain's properties and capabilities.Subjects--Topical Terms:

769149
Artificial Intelligence.

Neural networks in dynamical systems: A system theoretic approach.
LDR:02721nam 2200265 a 45 001 930028
005 20110427
008 110427s1992 eng d
035 $a (UnM)AAI9315226
035 $a AAI9315226
040 $a UnM $c UnM
100 1 $a Levin, Asriel Uzi. $3 1253511
245 1 0 $a Neural networks in dynamical systems: A system theoretic approach.
300 $a 150 p.
500 $a Source: Dissertation Abstracts International, Volume: 54-01, Section: B, page: 0468.
502 $a Thesis (Ph.D.)--Yale University, 1992.
520 $a Artificial neural networks is a rapidly developing field of research which concerns the study of massively parallel structures composed of simple non-linear processors that are interconnected via a set of modifiable connections. The motivation behind the study of these structures is based on empirical evidence that has been collected through the years about the organization of biological nervous systems. By building networks in whose structure we incorporate basic features of the biological nervous system, we hope to be able to synthesize artificial systems that exhibit some of the brain's properties and capabilities.
520 $a This dissertation is devoted to the study of neural network for the identification and control of nonlinear dynamical systems. Using the framework of system theory, we address the problem of control of unknown systems in a gradual and systematic fashion so as to isolate its different ingredients: regulation when the state of the system is accessible, state observation, identification, and finally control using only input-output observations. In order to make the problem tractable, certain assumptions concerning the controllability and observability properties of the system and its range of operation are made. In many cases, by confining attention to a neighborhood of an equilibrium state, the properties of the linearized system are shown to hold locally for the nonlinear system.
520 $a Throughout the thesis, a top-down approach is adopted. Each of the problems is first theoretically analyzed, hoping by that to understand the fundamental limitations that mathematics imposes on what is achievable irrespective of the particular technology that is used for implementation. Then, the theoretical analysis is used to determine the structure and training objectives of the neural networks needed to achieve the control goals. Finally, the effectiveness of the proposed methods is demonstrated through simulations.
590 $a School code: 0265.
650 4 $a Artificial Intelligence. $3 769149
650 4 $a Engineering, System Science. $3 1018128
690 $a 0790
690 $a 0800
710 2 0 $a Yale University. $3 515640
773 0 $t Dissertation Abstracts International $g 54-01B.
790 $a 0265
791 $a Ph.D.
792 $a 1992
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9315226