Value: $102.50 - $94.02
(as of Jul 13, 2025 04:30:01 UTC – Particulars)
This e book covers each classical and trendy fashions in deep studying. The first focus is on the idea and algorithms of deep studying. The idea and algorithms of neural networks are significantly necessary for understanding necessary ideas, in order that one can perceive the necessary design ideas of neural architectures in numerous functions. Why do neural networks work? When do they work higher than off-the-shelf machine-learning fashions? When is depth helpful? Why is coaching neural networks so onerous? What are the pitfalls? The e book can be wealthy in discussing totally different functions with the intention to give the practitioner a taste of how neural architectures are designed for several types of issues. Deep studying strategies for varied information domains, similar to textual content, photos, and graphs are introduced intimately. The chapters of this e book span three classes: The fundamentals of neural networks: The backpropagation algorithm is mentioned in Chapter 2.Many conventional machine studying fashions may be understood as particular circumstances of neural networks. Chapter 3 explores the connections between conventional machine studying and neural networks. Assist vector machines, linear/logistic regression, singular worth decomposition, matrix factorization, and recommender techniques are proven to be particular circumstances of neural networks. Fundamentals of neural networks: An in depth dialogue of coaching and regularization is supplied in Chapters 4 and 5. Chapters 6 and seven current radial-basis perform (RBF) networks and restricted Boltzmann machines. Superior matters in neural networks: Chapters 8, 9, and 10 talk about recurrent neural networks, convolutional neural networks, and graph neural networks. A number of superior matters like deep reinforcement studying, consideration mechanisms, transformer networks, Kohonen self-organizing maps, and generative adversarial networks are launched in Chapters 11 and 12. The textbook is written for graduate college students and higher below graduate degree college students. Researchers and practitioners working inside this associated area will need to buy this as properly.The place attainable, an application-centric view is highlighted with the intention to present an understanding of the sensible makes use of of every class of strategies.The second version is considerably reorganized and expanded with separate chapters on backpropagation and graph neural networks. Many chapters have been considerably revised over the primary version.Better focus is positioned on trendy deep studying concepts similar to consideration mechanisms, transformers, and pre-trained language fashions.
Writer : Springer Nature
Publication date : June 30 2023
Version : 2nd
Language : English
Print size : 529 pages
ISBN-10 : 3031296419
ISBN-13 : 978-3031296413
Merchandise weight : 215 g
Dimensions : 26.2 x 3.8 x 18.4 cm
