國立高雄大學圖資館 |

Language: English

Back

Algorithmic high-dimensional robust statistics

Record Type:	Electronic resources : Monograph/item
Title/Author:	Algorithmic high-dimensional robust statisticsIlias Diakonikolas, Daniel M. Kane.
Author:	Diakonikolas, Ilias.
other author:	Kane, Daniel M.,
Published:	Cambridge, United Kingdom ;Cambridge University Press,2023.
Description:	xvi, 283 p. :ill., digital ;23 cm.
Notes:	Title from publisher's bibliographic system (viewed on 30 Aug 2023).
Subject:	Robust statisticsData processing.
Online resource:	https://doi.org/10.1017/9781108943161
ISBN:	9781108943161

Algorithmic high-dimensional robust statistics
Diakonikolas, Ilias.

Algorithmic high-dimensional robust statistics[electronic resource] /Ilias Diakonikolas, Daniel M. Kane. - Cambridge, United Kingdom ;Cambridge University Press,2023. - xvi, 283 p. :ill., digital ;23 cm.

Title from publisher's bibliographic system (viewed on 30 Aug 2023).

Introduction to robust statistics -- Efficient high-dimensional robust mean estimation -- Algorithmic refinements in robust mean estimation -- Robust covariance estimation -- List-decodable learning -- Robust estimation via higher moments -- Robust supervised learning -- Information-computation trade-offs in high-dimensional robust statistics.

Robust statistics is the study of designing estimators that perform well even when the dataset significantly deviates from the idealized modeling assumptions, such as in the presence of model misspecification or adversarial outliers in the dataset. The classical statistical theory, dating back to pioneering works by Tukey and Huber, characterizes the information-theoretic limits of robust estimation for most common problems. A recent line of work in computer science gave the first computationally efficient robust estimators in high dimensions for a range of learning tasks. This reference text for graduate students, researchers, and professionals in machine learning theory, provides an overview of recent developments in algorithmic high-dimensional robust statistics, presenting the underlying ideas in a clear and unified manner, while leveraging new perspectives on the developed techniques to provide streamlined proofs of these results. The most basic and illustrative results are analyzed in each chapter, while more tangential developments are explored in the exercises.

ISBN: 9781108943161Subjects--Topical Terms:

973666
Robust statistics
--Data processing.

LC Class. No.: QA276.4 / .D53 2023

Dewey Class. No.: 519.502855369

Algorithmic high-dimensional robust statistics
LDR:02322nmm a2200265 a 4500 001 662046
003 UkCbUP
005 20230830022204.0
006 m d
007 cr nn 008maaau
008 241120s2023 enk o 1 0 eng d
020 $a 9781108943161 $q (electronic bk.)
020 $a 9781108837811 $q (hardback)
035 $a CR9781108943161
040 $a UkCbUP $b eng $c UkCbUP $d GP
041 0 $a eng
050 0 0 $a QA276.4 $b .D53 2023
082 0 4 $a 519.502855369 $2 23
090 $a QA276.4 $b .D536 2023
100 1 $a Diakonikolas, Ilias. $3 973664
245 1 0 $a Algorithmic high-dimensional robust statistics $h [electronic resource] / $c Ilias Diakonikolas, Daniel M. Kane.
260 $a Cambridge, United Kingdom ; $a New York, NY : $b Cambridge University Press, $c 2023.
300 $a xvi, 283 p. : $b ill., digital ; $c 23 cm.
500 $a Title from publisher's bibliographic system (viewed on 30 Aug 2023).
505 0 $a Introduction to robust statistics -- Efficient high-dimensional robust mean estimation -- Algorithmic refinements in robust mean estimation -- Robust covariance estimation -- List-decodable learning -- Robust estimation via higher moments -- Robust supervised learning -- Information-computation trade-offs in high-dimensional robust statistics.
520 $a Robust statistics is the study of designing estimators that perform well even when the dataset significantly deviates from the idealized modeling assumptions, such as in the presence of model misspecification or adversarial outliers in the dataset. The classical statistical theory, dating back to pioneering works by Tukey and Huber, characterizes the information-theoretic limits of robust estimation for most common problems. A recent line of work in computer science gave the first computationally efficient robust estimators in high dimensions for a range of learning tasks. This reference text for graduate students, researchers, and professionals in machine learning theory, provides an overview of recent developments in algorithmic high-dimensional robust statistics, presenting the underlying ideas in a clear and unified manner, while leveraging new perspectives on the developed techniques to provide streamlined proofs of these results. The most basic and illustrative results are analyzed in each chapter, while more tangential developments are explored in the exercises.
650 0 $a Robust statistics $x Data processing. $3 973666
650 0 $a Computer algorithms. $3 184478
700 1 $a Kane, Daniel M., $d 1986- $3 973665
856 4 0 $u https://doi.org/10.1017/9781108943161