國立虎尾科技大學 |

Data management in machine learning systems /

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Data management in machine learning systems // Matthias Boehm, Arun Kumar, Jun Yang.
作者:	Boehm, Matthias,
其他作者:	Kumar, Arun,
面頁冊數:	1 PDF (xv, 157 pages) :illustrations. :
附註:	Part of: Synthesis digital library of engineering and computer science.
標題:	Machine learning. -
電子資源:	https://ieeexplore.ieee.org/servlet/opac?bknumber=8653550
ISBN:	9781681734972

Data management in machine learning systems /
Boehm, Matthias,

Data management in machine learning systems /Matthias Boehm, Arun Kumar, Jun Yang. - 1 PDF (xv, 157 pages) :illustrations. - Synthesis lectures on data management,# 572153-5426 ;. - Synthesis digital library of engineering and computer science..

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

Includes bibliographical references (pages 127-156).

1. Introduction -- 1.1 Overview of ML lifecycle and ML users -- 1.2 Motivation -- 1.3 Outline and scope --

Abstract freely available; full-text restricted to subscribers or individual document purchasers.

Compendex

Large-scale data analytics using machine learning (ML) underpins many modern data-driven applications. ML systems provide means of specifying and executing these ML workloads in an efficient and scalable manner. Data management is at the heart of many ML systems due to data-driven application characteristics, data-centric workload characteristics, and system architectures inspired by classical data management techniques. In this book, we follow this data-centric view of ML systems and aim to provide a comprehensive overview of data management in ML systems for the end-to-end data science or ML lifecycle. We review multiple interconnected lines of work: (1) ML support in database (DB) systems, (2) DB-inspired ML systems, and (3) ML lifecycle systems. Covered topics include: in-database analytics via query generation and user-defined functions, factorized and statistical-relational learning; optimizing compilers for ML workloads; execution strategies and hardware accelerators; data access methods such as compression, partitioning and indexing; resource elasticity and cloud markets; as well as systems for data preparation for ML, model selection, model management, model debugging, and model serving. Given the rapidly evolving field, we strive for a balance between an up-to-date survey of ML systems, an overview of the underlying concepts and techniques, as well as pointers to open research questions. Hence, this book might serve as a starting point for both systems researchers and developers.

Mode of access: World Wide Web.

ISBN: 9781681734972

Standard No.: 10.2200/S00895ED1V01Y201901DTM057doiSubjects--Topical Terms:

561253
Machine learning.
Subjects--Index Terms:

ML systems

LC Class. No.: Q325.5 / .B643 2019

Dewey Class. No.: 006.31

Data management in machine learning systems /
LDR:05592nam 2200757 i 4500 001 959751
003 IEEE
005 20190227152155.0
006 m eo d
007 cr cn |||m|||a
008 201209s2019 caua foab 000 0 eng d
020 $a 9781681734972 $q ebook
020 $z 9781681734989 $q hardcover
020 $z 9781681734965 $q paperback
024 7 $a 10.2200/S00895ED1V01Y201901DTM057 $2 doi
035 $a (CaBNVSL)swl000409040
035 $a (OCoLC)1088564405
035 $a 8653550
040 $a CaBNVSL $b eng $e rda $c CaBNVSL $d CaBNVSL
050 4 $a Q325.5 $b .B643 2019
082 0 4 $a 006.31 $2 23
100 1 $a Boehm, Matthias, $e author. $3 1253065
245 1 0 $a Data management in machine learning systems / $c Matthias Boehm, Arun Kumar, Jun Yang.
264 1 $a [San Rafael, California] : $b Morgan & Claypool, $c 2019.
300 $a 1 PDF (xv, 157 pages) : $b illustrations.
336 $a text $2 rdacontent
337 $a electronic $2 isbdmedia
338 $a online resource $2 rdacarrier
490 1 $a Synthesis lectures on data management, $x 2153-5426 ; $v # 57
500 $a Part of: Synthesis digital library of engineering and computer science.
504 $a Includes bibliographical references (pages 127-156).
505 0 $a 1. Introduction -- 1.1 Overview of ML lifecycle and ML users -- 1.2 Motivation -- 1.3 Outline and scope --
505 8 $a 2. ML through database queries and UDFs -- 2.1 Linear algebra -- 2.2 Iterative algorithms -- 2.3 Sampling-based methods -- 2.4 Discussion -- 2.5 Summary --
505 8 $a 3. Multi-table ML and deep systems integration -- 3.1 Learning over joins -- 3.2 Statistical relational learning and non-IID models -- 3.3 Deeper integration and specialized DBMSs -- 3.4 Summary --
505 8 $a 4. Rewrites and optimization -- 4.1 Optimization scope -- 4.2 Logical rewrites and planning -- 4.3 Physical rewrites and operators -- 4.4 Automatic operator fusion -- 4.5 Runtime adaptation -- 4.6 Summary --
505 8 $a 5. Execution strategies -- 5.1 Data-parallel execution -- 5.2 Task-parallel execution -- 5.3 Parameter servers (model-parallel execution) -- 5.4 Hybrid execution strategies -- 5.5 Accelerators (GPUs, FPGAs, ASICs) -- 5.6 Summary --
505 8 $a 6. Data access methods -- 6.1 Caching and buffer pool management -- 6.2 Compression -- 6.3 NUMA-aware partitioning and replication -- 6.4 Index structures -- 6.5 Summary --
505 8 $a 7. Resource heterogeneity and elasticity -- 7.1 Provisioning, configuration, and scheduling -- 7.2 Handling failures -- 7.3 Working with markets of transient resources -- 7.4 Summary --
505 8 $a 8. Systems for ML lifecycle tasks -- 8.1 Data sourcing and cleaning for ML -- 8.2 Feature engineering and deep learning -- 8.3 Model selection and model management -- 8.4 Interaction, visualization, debugging, and inspection -- 8.5 Model deployment and serving -- 8.6 Benchmarking ML systems -- 8.7 Summary --
505 8 $a 9. Conclusions -- Bibliography -- Authors' biographies.
506 $a Abstract freely available; full-text restricted to subscribers or individual document purchasers.
510 0 $a Compendex
510 0 $a INSPEC
510 0 $a Google scholar
510 0 $a Google book search
520 3 $a Large-scale data analytics using machine learning (ML) underpins many modern data-driven applications. ML systems provide means of specifying and executing these ML workloads in an efficient and scalable manner. Data management is at the heart of many ML systems due to data-driven application characteristics, data-centric workload characteristics, and system architectures inspired by classical data management techniques. In this book, we follow this data-centric view of ML systems and aim to provide a comprehensive overview of data management in ML systems for the end-to-end data science or ML lifecycle. We review multiple interconnected lines of work: (1) ML support in database (DB) systems, (2) DB-inspired ML systems, and (3) ML lifecycle systems. Covered topics include: in-database analytics via query generation and user-defined functions, factorized and statistical-relational learning; optimizing compilers for ML workloads; execution strategies and hardware accelerators; data access methods such as compression, partitioning and indexing; resource elasticity and cloud markets; as well as systems for data preparation for ML, model selection, model management, model debugging, and model serving. Given the rapidly evolving field, we strive for a balance between an up-to-date survey of ML systems, an overview of the underlying concepts and techniques, as well as pointers to open research questions. Hence, this book might serve as a starting point for both systems researchers and developers.
530 $a Also available in print.
538 $a Mode of access: World Wide Web.
538 $a System requirements: Adobe Acrobat Reader.
588 $a Title from PDF title page (viewed on February 27, 2019).
650 0 $a Machine learning. $3 561253
650 0 $a Database management. $3 557799
653 $a ML systems
653 $a data management
653 $a data science
653 $a ML lifecycle
653 $a ML training
653 $a ML serving
653 $a in-database analytics
653 $a linear algebra
653 $a optimizing compilers
653 $a distributed machine learning
653 $a hardware accelerators
653 $a data access methods
653 $a resource elasticity
653 $a data cleaning
653 $a model management
653 $a benchmarking
700 1 $a Kumar, Arun, $e author. $3 1253066
700 1 $a Yang, Jun, $d 1975 September 9-, $e author. $3 1253067
776 0 8 $i Print version: $z 9781681734989 $z 9781681734965
830 0 $a Synthesis digital library of engineering and computer science. $3 598254
830 0 $a Synthesis lectures on data management ; $v #36. $3 931356
856 4 2 $3 Abstract with links to resource $u https://ieeexplore.ieee.org/servlet/opac?bknumber=8653550