國立虎尾科技大學 |

Foundations of deep learning

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Foundations of deep learning/ by Fengxiang He, Dacheng Tao.
作者:	He, Fengxiang.
其他作者:	Tao, Dacheng.
出版者:	Singapore :Springer Nature Singapore : : 2025.,
面頁冊數:	xiv, 292 p. :ill., digital ; : 24 cm.;
Contained By:	Springer Nature eBook
標題:	Artificial intelligence. -
電子資源:	https://doi.org/10.1007/978-981-16-8233-9
ISBN:	9789811682339

Foundations of deep learning
He, Fengxiang.

Foundations of deep learning[electronic resource] /by Fengxiang He, Dacheng Tao. - Singapore :Springer Nature Singapore :2025. - xiv, 292 p. :ill., digital ;24 cm. - Machine learning: foundations, methodologies, and applications,2730-9916. - Machine learning: foundations, methodologies, and applications..

Introduction -- Background -- Conventional Statistical Learning Theory -- Difficulty of Conventional Statistical Learning Theory -- Developing Deep Learning Theory -- Generalization Bounds on Hypothesis Complexity -- Interplay of Optimization, Bayesian Inference, and Generalization -- Geometrical Properties of Loss Surface -- The Role of Over-parametrization -- Rising Concerns in Ethics and Security -- Privacy Preservation -- Fairness Protection -- Algorithmic Robustness.

Deep learning has significantly reshaped a variety of technologies, such as image processing, natural language processing, and audio processing. The excellent generalizability of deep learning is like a "cloud" to conventional complexity-based learning theory: the over-parameterization of deep learning makes almost all existing tools vacuous. This irreconciliation considerably undermines the confidence of deploying deep learning to security-critical areas, including autonomous vehicles and medical diagnosis, where small algorithmic mistakes can lead to fatal disasters. This book seeks to explaining the excellent generalizability, including generalization analysis via the size-independent complexity measures, the role of optimization in understanding the generalizability, and the relationship between generalizability and ethical/security issues. The efforts to understand the excellent generalizability are following two major paths: (1) developing size-independent complexity measures, which can evaluate the "effective" hypothesis complexity that can be learned, instead of the whole hypothesis space; and (2) modelling the learned hypothesis through stochastic gradient methods, the dominant optimizers in deep learning, via stochastic differential functions and the geometry of the associated loss functions. Related works discover that over-parameterization surprisingly bring many good properties to the loss functions. Rising concerns of deep learning are seen on the ethical and security issues, including privacy preservation and adversarial robustness. Related works also reveal an interplay between them and generalizability: a good generalizability usually means a good privacy-preserving ability; and more robust algorithms might have a worse generalizability. We expect readers can have a big picture of the current knowledge in deep learning theory, understand how the deep learning theory can guide new algorithm designing, and identify future research directions. Readers need knowledge of calculus, linear algebra, probability, statistics, and statistical learning theory.

ISBN: 9789811682339

Standard No.: 10.1007/978-981-16-8233-9doiSubjects--Topical Terms:

559380
Artificial intelligence.

LC Class. No.: Q325.6

Dewey Class. No.: 006.31

Foundations of deep learning
LDR:03630nam a22003375a 4500 001 1172156
003 DE-He213
005 20250730164240.0
006 m d
007 cr nn 008maaau
008 260625s2025 si s 0 eng d
020 $a 9789811682339 $q (electronic bk.)
020 $a 9789811682322 $q (paper)
024 7 $a 10.1007/978-981-16-8233-9 $2 doi
035 $a 978-981-16-8233-9
040 $a GP $c GP
041 0 $a eng
050 4 $a Q325.6
072 7 $a UYQ $2 bicssc
072 7 $a COM004000 $2 bisacsh
072 7 $a UYQ $2 thema
082 0 4 $a 006.31 $2 23
090 $a Q325.6 $b .H432 2025
100 1 $a He, Fengxiang. $3 1502934
245 1 0 $a Foundations of deep learning $h [electronic resource] / $c by Fengxiang He, Dacheng Tao.
260 $a Singapore : $c 2025. $b Springer Nature Singapore : $b Imprint: Springer,
300 $a xiv, 292 p. : $b ill., digital ; $c 24 cm.
490 1 $a Machine learning: foundations, methodologies, and applications, $x 2730-9916
505 0 $a Introduction -- Background -- Conventional Statistical Learning Theory -- Difficulty of Conventional Statistical Learning Theory -- Developing Deep Learning Theory -- Generalization Bounds on Hypothesis Complexity -- Interplay of Optimization, Bayesian Inference, and Generalization -- Geometrical Properties of Loss Surface -- The Role of Over-parametrization -- Rising Concerns in Ethics and Security -- Privacy Preservation -- Fairness Protection -- Algorithmic Robustness.
520 $a Deep learning has significantly reshaped a variety of technologies, such as image processing, natural language processing, and audio processing. The excellent generalizability of deep learning is like a "cloud" to conventional complexity-based learning theory: the over-parameterization of deep learning makes almost all existing tools vacuous. This irreconciliation considerably undermines the confidence of deploying deep learning to security-critical areas, including autonomous vehicles and medical diagnosis, where small algorithmic mistakes can lead to fatal disasters. This book seeks to explaining the excellent generalizability, including generalization analysis via the size-independent complexity measures, the role of optimization in understanding the generalizability, and the relationship between generalizability and ethical/security issues. The efforts to understand the excellent generalizability are following two major paths: (1) developing size-independent complexity measures, which can evaluate the "effective" hypothesis complexity that can be learned, instead of the whole hypothesis space; and (2) modelling the learned hypothesis through stochastic gradient methods, the dominant optimizers in deep learning, via stochastic differential functions and the geometry of the associated loss functions. Related works discover that over-parameterization surprisingly bring many good properties to the loss functions. Rising concerns of deep learning are seen on the ethical and security issues, including privacy preservation and adversarial robustness. Related works also reveal an interplay between them and generalizability: a good generalizability usually means a good privacy-preserving ability; and more robust algorithms might have a worse generalizability. We expect readers can have a big picture of the current knowledge in deep learning theory, understand how the deep learning theory can guide new algorithm designing, and identify future research directions. Readers need knowledge of calculus, linear algebra, probability, statistics, and statistical learning theory.
650 0 $a Artificial intelligence. $3 559380
650 0 $a Machine learning. $3 561253
650 0 $a Neural networks (Computer science) $3 528588
650 0 $a Computer science $x Mathematics. $3 528496
650 0 $a Computational intelligence. $3 568984
650 1 4 $a Artificial Intelligence. $3 646849
650 2 4 $a Machine Learning. $3 1137723
650 2 4 $a Statistical Learning. $3 1396099
650 2 4 $a Mathematical Models of Cognitive Processes and Neural Networks. $3 884110
650 2 4 $a Mathematical Applications in Computer Science. $3 815331
650 2 4 $a Computational Intelligence. $3 768837
700 1 $a Tao, Dacheng. $e editor. $3 1256756
710 2 $a SpringerLink (Online service) $3 593884
773 0 $t Springer Nature eBook
830 0 $a Machine learning: foundations, methodologies, and applications. $3 1383597
856 4 0 $u https://doi.org/10.1007/978-981-16-8233-9
950 $a Artificial Intelligence (R0) (SpringerNature-85269)