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

Linked to FindBook

Google Book

Amazon

博客來

Fundamental Performance Limits of Statistical Problems : = From Detection Theory to Semi-Supervised Learning.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Fundamental Performance Limits of Statistical Problems :/
Reminder of title:	From Detection Theory to Semi-Supervised Learning.
Author:	He, Haiyun.
Description:	1 online resource (208 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 84-09, Section: B.
Contained By:	Dissertations Abstracts International84-09B.
Subject:	Cameras. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30340203click for full text (PQDT)
ISBN:	9798374490015

Fundamental Performance Limits of Statistical Problems : = From Detection Theory to Semi-Supervised Learning.
He, Haiyun.

Fundamental Performance Limits of Statistical Problems :From Detection Theory to Semi-Supervised Learning. - 1 online resource (208 pages)

Source: Dissertations Abstracts International, Volume: 84-09, Section: B.

Thesis (Ph.D.)--National University of Singapore (Singapore), 2022.

Includes bibliographical references

Studying and designing close-to-optimal mechanisms to infer or learn useful information from raw data is of tremendous significance in this digital era. This thesis explores the fundamental performance limits of three classes of statistical problems-distributed detection, change-point detection and the generalization capabilities of semi-supervised learning (SSL). In a sensor network, the distributed detection problem concerns the scenario in which the fusion center needs to make a decision based on data sent from a number of sensors via different channels. The change-point detection problem concerns the scenario in which the data samples are collected under different conditions and one needs to estimate the change-points of the condition. In contrast to classical works where the underlying data distributions are assumed to be known, we consider the practical scenario where training data samples are available instead. For distributed detection, we derive the asymptotically optimal type-II error exponent given that the type-I error decays exponentially fast. We also derive the asymptotically optimal test at the fusion center. For change-point detection, we derive the asymptotically optimal change-point estimator in both large and moderate deviations regimes, as well as the asymptotically optimal detection confidence width as a function of the undetected error. Finally, we consider a more complicated scenario in which we are motivated to mitigate the high cost of labelling data. To do so, we analyse the fundamental limits of SSL which makes use of both labelled and unlabelled data. Using information-theoretic principles, we investigate the generalization performance of SSL, which quantifies the extent to which the algorithms overfits to the training data. We show that under iterative SSL with pseudo-labelling, for easier-to-distinguish classes, the generalization error decreases rapidly in the first few iterations and saturates afterwards while for difficult-todistinguish classes, the generalization error increases instead. Regularization can help to mitigate this undesirable effect. Our experiments on benchmark datasets such as the MNIST and CIFAR-10 datasets corroborate our theoretical results.

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

Mode of access: World Wide Web

ISBN: 9798374490015Subjects--Topical Terms:

524039
Cameras.
Index Terms--Genre/Form:

542853
Electronic books.

Fundamental Performance Limits of Statistical Problems : = From Detection Theory to Semi-Supervised Learning.
LDR:03563nmm a2200361K 4500 001 2360270
005 20230926101828.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798374490015
035 $a (MiAaPQ)AAI30340203
035 $a (MiAaPQ)USingapore231542
035 $a AAI30340203
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a He, Haiyun. $3 3700885
245 1 0 $a Fundamental Performance Limits of Statistical Problems : $b From Detection Theory to Semi-Supervised Learning.
264 0 $c 2022
300 $a 1 online resource (208 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: 84-09, Section: B.
500 $a Advisor: Tan, Vincent Yan Fu.
502 $a Thesis (Ph.D.)--National University of Singapore (Singapore), 2022.
504 $a Includes bibliographical references
520 $a Studying and designing close-to-optimal mechanisms to infer or learn useful information from raw data is of tremendous significance in this digital era. This thesis explores the fundamental performance limits of three classes of statistical problems-distributed detection, change-point detection and the generalization capabilities of semi-supervised learning (SSL). In a sensor network, the distributed detection problem concerns the scenario in which the fusion center needs to make a decision based on data sent from a number of sensors via different channels. The change-point detection problem concerns the scenario in which the data samples are collected under different conditions and one needs to estimate the change-points of the condition. In contrast to classical works where the underlying data distributions are assumed to be known, we consider the practical scenario where training data samples are available instead. For distributed detection, we derive the asymptotically optimal type-II error exponent given that the type-I error decays exponentially fast. We also derive the asymptotically optimal test at the fusion center. For change-point detection, we derive the asymptotically optimal change-point estimator in both large and moderate deviations regimes, as well as the asymptotically optimal detection confidence width as a function of the undetected error. Finally, we consider a more complicated scenario in which we are motivated to mitigate the high cost of labelling data. To do so, we analyse the fundamental limits of SSL which makes use of both labelled and unlabelled data. Using information-theoretic principles, we investigate the generalization performance of SSL, which quantifies the extent to which the algorithms overfits to the training data. We show that under iterative SSL with pseudo-labelling, for easier-to-distinguish classes, the generalization error decreases rapidly in the first few iterations and saturates afterwards while for difficult-todistinguish classes, the generalization error increases instead. Regularization can help to mitigate this undesirable effect. Our experiments on benchmark datasets such as the MNIST and CIFAR-10 datasets corroborate our theoretical results.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Cameras. $3 524039
650 4 $a Smart houses. $3 3700886
650 4 $a Algorithms. $3 536374
650 4 $a Information theory. $3 542527
650 4 $a Sensors. $3 3549539
650 4 $a Labeling. $3 3560710
650 4 $a Applied mathematics. $3 2122814
650 4 $a Mathematics. $3 515831
650 4 $a Statistics. $3 517247
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0364
690 $a 0800
690 $a 0405
690 $a 0463
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a National University of Singapore (Singapore). $3 3352228
773 0 $t Dissertations Abstracts International $g 84-09B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30340203 $z click for full text (PQDT)