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

FindBook

Google Book

Amazon

博客來

Scaling up Recognition in Expert Domains with Crowd-Source Annotations.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Scaling up Recognition in Expert Domains with Crowd-Source Annotations./
作者:	Wang, Pei.
面頁冊數:	1 online resource (174 pages)
附註:	Source: Dissertations Abstracts International, Volume: 84-07, Section: A.
Contained By:	Dissertations Abstracts International84-07A.
標題:	Electrical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29398358click for full text (PQDT)
ISBN:	9798368442457

Scaling up Recognition in Expert Domains with Crowd-Source Annotations.
Wang, Pei.

Scaling up Recognition in Expert Domains with Crowd-Source Annotations. - 1 online resource (174 pages)

Source: Dissertations Abstracts International, Volume: 84-07, Section: A.

Thesis (Ph.D.)--University of California, San Diego, 2022.

Includes bibliographical references

The success of deep learning in image recognition is substantially driven by large-scale, well-curated data. On visual recognition of common objects, the data can be scalably annotated on online crowd-sourcing platforms because the labeling does not need any prior knowledge. However, the case is not true for images of expertise like biological or medical imaging in which labeling them needs background knowledge. Although data collection is still usually easy, the annotation is difficult. Existing self-supervised or semi-supervised solutions train a model that tries to learn from a small amount of labeled data and a large amount of unlabeled data. These solutions show good performances on common object recognition but have been found not to work effectively on fine-grained expert domains.In this thesis, we propose a new solution with crowd source annotations to address the problem. Inspired by the fact that supervised learning on as much as data can always perform better, our method tries to scale up the annotation. This is implemented by two different approaches, machine teaching and human filtering. Machine teaching first teaches humans with a short carefully designed course to learn the expertise knowledge so that they can label the data later. Human filtering simplifies the process to a binary selection procedure without preceding training. Beyond these two approaches, a unified explanation framework is developed to generate visualizations that are merged into two approaches, enabling easier and more accurate annotation results. Experiments show that both methods significantly outperform various alternative approaches in several benchmarks. They have also been found to be versatile and can benefit from more advanced machine learning techniques in the future. Overall, we believe that this thesis opens up a new direction to think about the expert domain classification problem, in general.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798368442457Subjects--Topical Terms:

649834
Electrical engineering.
Subjects--Index Terms:

Deep learningIndex Terms--Genre/Form:

542853
Electronic books.

Scaling up Recognition in Expert Domains with Crowd-Source Annotations.
LDR:03231nmm a2200373K 4500 001 2362351
005 20231027104010.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798368442457
035 $a (MiAaPQ)AAI29398358
035 $a AAI29398358
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Wang, Pei. $3 1620243
245 1 0 $a Scaling up Recognition in Expert Domains with Crowd-Source Annotations.
264 0 $c 2022
300 $a 1 online resource (174 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 84-07, Section: A.
500 $a Advisor: Vasconcelos, Nuno.
502 $a Thesis (Ph.D.)--University of California, San Diego, 2022.
504 $a Includes bibliographical references
520 $a The success of deep learning in image recognition is substantially driven by large-scale, well-curated data. On visual recognition of common objects, the data can be scalably annotated on online crowd-sourcing platforms because the labeling does not need any prior knowledge. However, the case is not true for images of expertise like biological or medical imaging in which labeling them needs background knowledge. Although data collection is still usually easy, the annotation is difficult. Existing self-supervised or semi-supervised solutions train a model that tries to learn from a small amount of labeled data and a large amount of unlabeled data. These solutions show good performances on common object recognition but have been found not to work effectively on fine-grained expert domains.In this thesis, we propose a new solution with crowd source annotations to address the problem. Inspired by the fact that supervised learning on as much as data can always perform better, our method tries to scale up the annotation. This is implemented by two different approaches, machine teaching and human filtering. Machine teaching first teaches humans with a short carefully designed course to learn the expertise knowledge so that they can label the data later. Human filtering simplifies the process to a binary selection procedure without preceding training. Beyond these two approaches, a unified explanation framework is developed to generate visualizations that are merged into two approaches, enabling easier and more accurate annotation results. Experiments show that both methods significantly outperform various alternative approaches in several benchmarks. They have also been found to be versatile and can benefit from more advanced machine learning techniques in the future. Overall, we believe that this thesis opens up a new direction to think about the expert domain classification problem, in general.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Electrical engineering. $3 649834
650 4 $a Information science. $3 554358
653 $a Deep learning
653 $a Image recognition
653 $a Big data
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0544
690 $a 0723
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a University of California, San Diego. $b Electrical and Computer Engineering. $3 3432690
773 0 $t Dissertations Abstracts International $g 84-07A.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29398358 $z click for full text (PQDT)