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

Network Structures, Concurrency, and...

Cooper, Hal James.

FindBook

Google Book

Amazon

博客來

Network Structures, Concurrency, and Interpretability: Lessons from the Development of an AI Enabled Graph Database System.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Network Structures, Concurrency, and Interpretability: Lessons from the Development of an AI Enabled Graph Database System./
作者:	Cooper, Hal James.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	174 p.
附註:	Source: Dissertations Abstracts International, Volume: 81-05, Section: B.
Contained By:	Dissertations Abstracts International81-05B.
標題:	Computer science. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27544154
ISBN:	9781392874592

Network Structures, Concurrency, and Interpretability: Lessons from the Development of an AI Enabled Graph Database System.
Cooper, Hal James.

Network Structures, Concurrency, and Interpretability: Lessons from the Development of an AI Enabled Graph Database System. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 174 p.

Source: Dissertations Abstracts International, Volume: 81-05, Section: B.

Thesis (Ph.D.)--Columbia University, 2020.

This item must not be sold to any third party vendors.

This thesis describes the development of the SmartGraph, an AI enabled graph database.The need for such a system has been independently recognized in the isolated fields of graphdatabases, graph computing, and computational graph deep learning systems, such as TensorFlow.Though prior works have investigated some relationships between these fields, we believe that theSmartGraph is the first system designed from conception to incorporate the most significant anduseful characteristics of each. Examples include the ability to store graph structured data, runanalytics natively on this data, and run gradient descent algorithms. It is the synergistic aspectsof combining these fields that provide the most novel results presented in this dissertation. Keyamong them is how the notion of "graph querying" as used in graph databases can be used to solvea problem that has plagued deep learning systems since their inception; rather than attempting toembed graph structured datasets into restrictive vector spaces, we instead allow the deep learningfunctionality of the system to natively perform graph querying in memory during optimization as away of interpreting (and learning) the graph. This results in a concept of natural and interpretableprocessing of graph structured data.Graph computing systems have traditionally used distributed computing across multiple computenodes (e.g. separate machines connected via Ethernet or internet) to deal with large-scaledatasets whilst working sequentially on problems over entire datasets. In this dissertation, weoutline a distributed graph computing methodology that facilitates all the above capabilities (evenin an environment consisting of a single physical machine) while allowing for a workflow more typicalof a graph database than a graph computing system; massive concurrent access allowing forarbitrarily asynchronous execution of queries and analytics across the entire system. Further, wedemonstrate how this methodology is key to the artificial intelligence capabilities of the system.

ISBN: 9781392874592Subjects--Topical Terms:

523869
Computer science.

Network Structures, Concurrency, and Interpretability: Lessons from the Development of an AI Enabled Graph Database System.
LDR:03064nmm a2200301 4500 001 2264363
005 20200423113021.5
008 220629s2020 ||||||||||||||||| ||eng d
020 $a 9781392874592
035 $a (MiAaPQ)AAI27544154
035 $a AAI27544154
040 $a MiAaPQ $c MiAaPQ
100 1 $a Cooper, Hal James. $3 3541478
245 1 0 $a Network Structures, Concurrency, and Interpretability: Lessons from the Development of an AI Enabled Graph Database System.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 174 p.
500 $a Source: Dissertations Abstracts International, Volume: 81-05, Section: B.
500 $a Advisor: Iyengar, Garud N.
502 $a Thesis (Ph.D.)--Columbia University, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a This thesis describes the development of the SmartGraph, an AI enabled graph database.The need for such a system has been independently recognized in the isolated fields of graphdatabases, graph computing, and computational graph deep learning systems, such as TensorFlow.Though prior works have investigated some relationships between these fields, we believe that theSmartGraph is the first system designed from conception to incorporate the most significant anduseful characteristics of each. Examples include the ability to store graph structured data, runanalytics natively on this data, and run gradient descent algorithms. It is the synergistic aspectsof combining these fields that provide the most novel results presented in this dissertation. Keyamong them is how the notion of "graph querying" as used in graph databases can be used to solvea problem that has plagued deep learning systems since their inception; rather than attempting toembed graph structured datasets into restrictive vector spaces, we instead allow the deep learningfunctionality of the system to natively perform graph querying in memory during optimization as away of interpreting (and learning) the graph. This results in a concept of natural and interpretableprocessing of graph structured data.Graph computing systems have traditionally used distributed computing across multiple computenodes (e.g. separate machines connected via Ethernet or internet) to deal with large-scaledatasets whilst working sequentially on problems over entire datasets. In this dissertation, weoutline a distributed graph computing methodology that facilitates all the above capabilities (evenin an environment consisting of a single physical machine) while allowing for a workflow more typicalof a graph database than a graph computing system; massive concurrent access allowing forarbitrarily asynchronous execution of queries and analytics across the entire system. Further, wedemonstrate how this methodology is key to the artificial intelligence capabilities of the system.
590 $a School code: 0054.
650 4 $a Computer science. $3 523869
650 4 $a Operations research. $3 547123
690 $a 0984
690 $a 0796
710 2 $a Columbia University. $b Operations Research. $3 2096452
773 0 $t Dissertations Abstracts International $g 81-05B.
790 $a 0054
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27544154