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

Active learning

Settles, Burr.

Linked to FindBook

Google Book

Amazon

博客來

Active learning

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Active learning/ Burr Settles.
Author:	Settles, Burr.
Published:	[S.l.] :Morgan & Claypool, : cop. 2012.,
Description:	1 online resource (Text (xiii, 100 p.) :)ill.
Notes:	Part of: Synthesis digital library of engineering and computer science.
[NT 15003449]:	Preface -- Acknowledgments --
[NT 15003449]:	1. Automating inquiry -- 1.1 A thought experiment -- 1.2 Active learning -- 1.3 Scenarios for active learning --
[NT 15003449]:	2. Uncertainty sampling -- 2.1 Pushing the boundaries -- 2.2 An example -- 2.3 Measures of uncertainty -- 2.4 Beyond classification -- 2.5 Discussion --
[NT 15003449]:	3. Searching through the hypothesis space -- 3.1 The version space -- 3.2 Uncertainty sampling as version space search -- 3.3 Query by disagreement -- 3.4 Query by committee -- 3.5 Discussion --
[NT 15003449]:	4. Minimizing expected error and variance -- 4.1 Expected error reduction -- 4.2 Variance reduction -- 4.3 Batch queries and submodularity -- 4.4 Discussion --
[NT 15003449]:	5. Exploiting structure in data -- 5.1 Density-weighted methods -- 5.2 Cluster-based active learning -- 5.3 Active + semi-supervised learning -- 5.4 Discussion --
[NT 15003449]:	6. Theory -- 6.1 A unified view -- 6.2 A PAC bound for active learning -- 6.3 Discussion --
[NT 15003449]:	7. Practical considerations -- 7.1 Which algorithm is best? -- 7.2 Real labeling costs -- 7.3 Alternative query types -- 7.4 Skewed label distributions -- 7.5 Unreliable oracles -- 7.6 Multi-task active learning -- 7.7 Data reuse and the unknown model class -- 7.8 Stopping criteria --
[NT 15003449]:	A. Nomenclature reference -- Bibliography -- Author's biography -- Index.
Subject:	Supervised learning (Machine learning) -
ISBN:	9781608457267 (ebk)

Active learning
Settles, Burr.

Active learning[elektronische middelen] /Burr Settles. - [S.l.] :Morgan & Claypool,cop. 2012. - 1 online resource (Text (xiii, 100 p.) :)ill. - Synthesis lectures on artificial intelligence and machine learning, ISSN 1939-4616 ;18.

Part of: Synthesis digital library of engineering and computer science.

Includes bibliographical references (p. 81-96) and index.

Preface -- Acknowledgments --

The key idea behind active learning is that a machine learning algorithm can perform better with less training if it is allowed to choose the data from which it learns. An active learner may pose "queries," usually in the form of unlabeled data instances to be labeled by an "oracle" (e.g., a human annotator) that already understands the nature of the problem. This sort of approach is well-motivated in many modern machine learning and data mining applications, where unlabeled data may be abundant or easy to come by, but training labels are difficult, time-consuming, or expensive to obtain. This book is a general introduction to active learning. It outlines several scenarios in which queries might be formulated, and details many query selection algorithms which have been organized into four broad categories, or "query selection frameworks."We also touch on some of the theoretical foundations of active learning, and conclude with an overview of the strengths and weaknesses of these approaches in practice, including a summary of ongoing work to address these open challenges and opportunities.

ISBN: 9781608457267 (ebk)Subjects--Topical Terms:

1534356
Supervised learning (Machine learning)
Subjects--Index Terms:

active learning

LC Class. No.: Q325.75 / .S472 2012

Active learning
LDR:03985nam a2200529Mi 4500 001 1264028
003 OCoLC
005 20120817033848.0
006 m o d u
007 cr |||||||||||
008 120817s2012 xx o 000 0 eng d
020 $a 9781608457267 (ebk)
020 $a 1608457265 (ebk)
020 $a 9781608457250 (pbk) $c US40
020 $a 1608457257 (pbk)
029 0 $a NLGGC $b 344538443
035 $a LIB-BW-101-015
035 $a (OCoLC)804676285
035 $a ocn804676285
040 $a NLGGC $e fobidrtb $b dut $c NLGGC
050 # 0 $a Q325.75 $b .S472 2012
100 1 $a Settles, Burr. $3 1530102
245 1 0 $a Active learning $h [elektronische middelen] / $c Burr Settles.
260 # $a [S.l.] : $b Morgan & Claypool, $c cop. 2012.
300 $a 1 online resource (Text (xiii, 100 p.) :) $b ill.
490 0 $a Synthesis lectures on artificial intelligence and machine learning, ISSN 1939-4616 ; $v 18
500 $a Part of: Synthesis digital library of engineering and computer science.
500 $a Title from PDF t.p. (Morgan & Claypool, viewed on July 14, 2012).
504 $a Includes bibliographical references (p. 81-96) and index.
504 $a Met lit. opg. en reg.
505 0 # $a Preface -- Acknowledgments --
505 8 # $a 1. Automating inquiry -- 1.1 A thought experiment -- 1.2 Active learning -- 1.3 Scenarios for active learning --
505 8 # $a 2. Uncertainty sampling -- 2.1 Pushing the boundaries -- 2.2 An example -- 2.3 Measures of uncertainty -- 2.4 Beyond classification -- 2.5 Discussion --
505 8 # $a 3. Searching through the hypothesis space -- 3.1 The version space -- 3.2 Uncertainty sampling as version space search -- 3.3 Query by disagreement -- 3.4 Query by committee -- 3.5 Discussion --
505 8 # $a 4. Minimizing expected error and variance -- 4.1 Expected error reduction -- 4.2 Variance reduction -- 4.3 Batch queries and submodularity -- 4.4 Discussion --
505 8 # $a 5. Exploiting structure in data -- 5.1 Density-weighted methods -- 5.2 Cluster-based active learning -- 5.3 Active + semi-supervised learning -- 5.4 Discussion --
505 8 # $a 6. Theory -- 6.1 A unified view -- 6.2 A PAC bound for active learning -- 6.3 Discussion --
505 8 # $a 7. Practical considerations -- 7.1 Which algorithm is best? -- 7.2 Real labeling costs -- 7.3 Alternative query types -- 7.4 Skewed label distributions -- 7.5 Unreliable oracles -- 7.6 Multi-task active learning -- 7.7 Data reuse and the unknown model class -- 7.8 Stopping criteria --
505 8 # $a A. Nomenclature reference -- Bibliography -- Author's biography -- Index.
520 3 $a The key idea behind active learning is that a machine learning algorithm can perform better with less training if it is allowed to choose the data from which it learns. An active learner may pose "queries," usually in the form of unlabeled data instances to be labeled by an "oracle" (e.g., a human annotator) that already understands the nature of the problem. This sort of approach is well-motivated in many modern machine learning and data mining applications, where unlabeled data may be abundant or easy to come by, but training labels are difficult, time-consuming, or expensive to obtain. This book is a general introduction to active learning. It outlines several scenarios in which queries might be formulated, and details many query selection algorithms which have been organized into four broad categories, or "query selection frameworks."We also touch on some of the theoretical foundations of active learning, and conclude with an overview of the strengths and weaknesses of these approaches in practice, including a summary of ongoing work to address these open challenges and opportunities.
650 0 0 $a Supervised learning (Machine learning) $3 1534356
650 0 0 $a Explanation-based learning. $3 1534357
650 # 0 $a COMPUTERS / Enterprise Applications / Business Intelligence Tools $2 bisacsh $3 1534358
650 # 0 $a COMPUTERS / Intelligence (AI) & Semantics $2 bisacsh $3 1534359
653 # # $a active learning
653 # # $a expected error reduction
653 # # $a hierarchical sampling
653 # # $a optimal experimental design
653 # # $a query by committee
653 # # $a query by disagreement
653 # # $a query learning
653 # # $a uncertainty sampling
653 # # $a variance reduction