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

Time-Ordered Online Training of Conv...

Teng, Ervin.

Linked to FindBook

Google Book

Amazon

博客來

Time-Ordered Online Training of Convolutional Neural Networks.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Time-Ordered Online Training of Convolutional Neural Networks./
Author:	Teng, Ervin.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
Description:	133 p.
Notes:	Source: Dissertations Abstracts International, Volume: 80-12, Section: B.
Contained By:	Dissertations Abstracts International80-12B.
Subject:	Computer Engineering. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13878476
ISBN:	9781392178430

Time-Ordered Online Training of Convolutional Neural Networks.
Teng, Ervin.

Time-Ordered Online Training of Convolutional Neural Networks. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 133 p.

Source: Dissertations Abstracts International, Volume: 80-12, Section: B.

Thesis (Ph.D.)--Carnegie Mellon University, 2019.

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

Learned algorithms, especially deep neural networks, have proliferated in recent years, primarily due to their adaptability and performance. For applications where humans are required to monitor a datastream, e.g. a video feed from a UAS, such algorithms can greatly reduce the cognitive load required of human operators. But these algorithms are learned, not learning. After deployment to an end device such as a UAS, they are generally unchanged until the next software update. This makes them inflexible to changing requirements and scenarios, a sharp contrast to adaptable human operators. Furthermore, retraining requires data collection and annotation-sometimes upwards of thousands of images-and must be done offline, not while a particular mission is unfolding.In this work, we tackle the problem of online, real-time training of a deep learning algorithm. We refer to this type of problem as time-ordered online training (ToOT), of which the UAS use case is an example. We begin with the observation that in human students, learning requires both study and curiosity. A good learner is not only good at extracting information from the data given to it, but also skilled at finding the right new information to learn from. They may have help from a mentor or teacher, but must use this assistance wisely and effectively. For a real-time learning algorithm onboard a UAS, this means being able to learn from an incoming video feed while minimizing the human annotations and time required to do so, and moving around the environment to augment its learning. We define a metric, incremental training benefit (ITB) per annotation, that seeks to capture the value extracted for each annotation provided by the user.We present a combined system that addresses both study and curiosity when deployed onboard a small UAS. This system has two components: ClickBAIT and autonomous curiosity. ClickBAIT is a framework designed to enable real-time training of a learning algorithm from mouse clicks on a video feed. We apply this framework to classification as well as detection models, showing that it increases the ITB of each annotation by 3-7 times across both cases. Furthermore, we show that such techniques can be used to enable learning from a separate low-power sensor network, without clicks at all. Autonomous curiosity is a system, trained through deep reinforcement learning, that accomplishes the task of both guiding the UAS' motion and choosing when to ask for annotations from the human operator. Our trained approach is able to increase average ITB by 10-14 times versus a random search. In addition, we examine how autonomous curiosity agents can be trained to consider the time-cost as well as the annotation cost of its actions, adapting its behavior to optimize for the characteristics of a particular UAS.

ISBN: 9781392178430Subjects--Topical Terms:

1567821
Computer Engineering.

Time-Ordered Online Training of Convolutional Neural Networks.
LDR:03900nmm a2200337 4500 001 2208556
005 20191021073502.5
008 201008s2019 ||||||||||||||||| ||eng d
020 $a 9781392178430
035 $a (MiAaPQ)AAI13878476
035 $a (MiAaPQ)cmu:10382
035 $a AAI13878476
040 $a MiAaPQ $c MiAaPQ
100 1 $a Teng, Ervin. $3 3435595
245 1 0 $a Time-Ordered Online Training of Convolutional Neural Networks.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 133 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-12, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Iannucci, Bob A.
502 $a Thesis (Ph.D.)--Carnegie Mellon University, 2019.
506 $a This item must not be sold to any third party vendors.
520 $a Learned algorithms, especially deep neural networks, have proliferated in recent years, primarily due to their adaptability and performance. For applications where humans are required to monitor a datastream, e.g. a video feed from a UAS, such algorithms can greatly reduce the cognitive load required of human operators. But these algorithms are learned, not learning. After deployment to an end device such as a UAS, they are generally unchanged until the next software update. This makes them inflexible to changing requirements and scenarios, a sharp contrast to adaptable human operators. Furthermore, retraining requires data collection and annotation-sometimes upwards of thousands of images-and must be done offline, not while a particular mission is unfolding.In this work, we tackle the problem of online, real-time training of a deep learning algorithm. We refer to this type of problem as time-ordered online training (ToOT), of which the UAS use case is an example. We begin with the observation that in human students, learning requires both study and curiosity. A good learner is not only good at extracting information from the data given to it, but also skilled at finding the right new information to learn from. They may have help from a mentor or teacher, but must use this assistance wisely and effectively. For a real-time learning algorithm onboard a UAS, this means being able to learn from an incoming video feed while minimizing the human annotations and time required to do so, and moving around the environment to augment its learning. We define a metric, incremental training benefit (ITB) per annotation, that seeks to capture the value extracted for each annotation provided by the user.We present a combined system that addresses both study and curiosity when deployed onboard a small UAS. This system has two components: ClickBAIT and autonomous curiosity. ClickBAIT is a framework designed to enable real-time training of a learning algorithm from mouse clicks on a video feed. We apply this framework to classification as well as detection models, showing that it increases the ITB of each annotation by 3-7 times across both cases. Furthermore, we show that such techniques can be used to enable learning from a separate low-power sensor network, without clicks at all. Autonomous curiosity is a system, trained through deep reinforcement learning, that accomplishes the task of both guiding the UAS' motion and choosing when to ask for annotations from the human operator. Our trained approach is able to increase average ITB by 10-14 times versus a random search. In addition, we examine how autonomous curiosity agents can be trained to consider the time-cost as well as the annotation cost of its actions, adapting its behavior to optimize for the characteristics of a particular UAS.
590 $a School code: 0041.
650 4 $a Computer Engineering. $3 1567821
650 4 $a Robotics. $3 519753
650 4 $a Computer science. $3 523869
690 $a 0464
690 $a 0771
690 $a 0984
710 2 $a Carnegie Mellon University. $b Electrical and Computer Engineering. $3 2094139
773 0 $t Dissertations Abstracts International $g 80-12B.
790 $a 0041
791 $a Ph.D.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13878476