東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Multiagent Data Fusion: Subspace Mod...

Li, Lin.

FindBook

Google Book

Amazon

博客來

Multiagent Data Fusion: Subspace Model Discovery and Learning via Network Diffusion.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Multiagent Data Fusion: Subspace Model Discovery and Learning via Network Diffusion./
作者:	Li, Lin.
面頁冊數:	158 p.
附註:	Source: Dissertation Abstracts International, Volume: 75-10(E), Section: B.
Contained By:	Dissertation Abstracts International75-10B(E).
標題:	Electrical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3626828
ISBN:	9781321018820

Multiagent Data Fusion: Subspace Model Discovery and Learning via Network Diffusion.
Li, Lin.

Multiagent Data Fusion: Subspace Model Discovery and Learning via Network Diffusion. - 158 p.

Source: Dissertation Abstracts International, Volume: 75-10(E), Section: B.

Thesis (Ph.D.)--University of California, Davis, 2013.

In today's highly computerized environment, large amounts of data are created every second. These unprocessed data can be from multiple information sources (e.g., mobile phones calls, SMS, GPS locator, emails, tweets, social networking sites, etc.) and/or from data sources that are distributed in nature (e.g., wireless sensor networks, distributed data mining systems, cloud computing systems, etc.). They are often high dimensional, scattered, and non-stationary, but the relevant information about the underlying processes of interest generally lies in a lower dimensional embedding space of a few parameters.

ISBN: 9781321018820Subjects--Topical Terms:

649834
Electrical engineering.

Multiagent Data Fusion: Subspace Model Discovery and Learning via Network Diffusion.
LDR:03141nmm a2200301 4500 001 2076086
005 20161101084235.5
008 170521s2013 ||||||||||||||||| ||eng d
020 $a 9781321018820
035 $a (MiAaPQ)AAI3626828
035 $a AAI3626828
040 $a MiAaPQ $c MiAaPQ
100 1 $a Li, Lin. $3 1020924
245 1 0 $a Multiagent Data Fusion: Subspace Model Discovery and Learning via Network Diffusion.
300 $a 158 p.
500 $a Source: Dissertation Abstracts International, Volume: 75-10(E), Section: B.
500 $a Adviser: Anna Scaglione.
502 $a Thesis (Ph.D.)--University of California, Davis, 2013.
520 $a In today's highly computerized environment, large amounts of data are created every second. These unprocessed data can be from multiple information sources (e.g., mobile phones calls, SMS, GPS locator, emails, tweets, social networking sites, etc.) and/or from data sources that are distributed in nature (e.g., wireless sensor networks, distributed data mining systems, cloud computing systems, etc.). They are often high dimensional, scattered, and non-stationary, but the relevant information about the underlying processes of interest generally lies in a lower dimensional embedding space of a few parameters.
520 $a My research goal is precisely to model and analyze large quantities of high-dimensional data that are collected from various sources in a multi-agent system, and to theoretically study the underlying diffusion patterns as the result of a complex web of interactions in a network of agents. To this end, this dissertation is divided into two parts. The first parts focuses on investigating methods for extracting low-dimensional structure from high-dimensional data and adaptively tracking the principal subspace of data streams that are collected at different agents. We also analyze the case of learning the data conveyed from a distributed network using an in-network gossip-based protocol. We show that relatively simple iterative message passing followed by local computations can converge to the optimum solution while reducing the computation burden that arises from centralized processing. This motivates the second part of this thesis, which focuses on the mathematical modeling and theoretical tools for analyzing information propagation and diffusion in networks. We compare models for social learning in the context of economics and social sciences with those used in the study of distributed sensor networks, explaining how consensus emerges or fails to emerge in both scenarios. Furthermore, we investigate various local interaction models in social networks with an emphasis on the situation where agents interact more strongly with the agents that have similar beliefs. The question that we seek to address is how social interactions lead to the formation of opinions in a social group.
590 $a School code: 0029.
650 4 $a Electrical engineering. $3 649834
650 4 $a Information technology. $3 532993
650 4 $a Engineering. $3 586835
690 $a 0544
690 $a 0489
690 $a 0537
710 2 $a University of California, Davis. $b Electrical and Computer Engineering. $3 1672487
773 0 $t Dissertation Abstracts International $g 75-10B(E).
790 $a 0029
791 $a Ph.D.
792 $a 2013
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3626828