Ye, Jong Chul
Geometry of Deep Learning: A Signal Processing Perspective (Mathematics in Industry)
- ISBN 13:
- 9789811660450
- author:
- Ye, Jong Chul
- format:
- Hardback
- publisher:
- Springer
- language:
- English
- Publication Year:
- 2022
- Pages:
- 330
- Dimensions:
- 23.4 x 15.6 x 2.1 centimetres (0
- Genre:
- Professional & Vocational, Computers, Artificial Intelligence,
- Condition:
- New
- Availability:
- Item usually sent within 5 working days
Description
Geometry of Deep Learning: A Signal Processing Perspective by Jong Chul Ye provides a unified understanding of deep learning through the lens of geometry. This book presents a signal processing perspective, explaining deep learning as an ultimate form of signal processing techniques. It delves into classical kernel machine learning approaches and their geometric structure, before exploring the latest tools in deep neural networks. The book offers a detailed analysis of the basic building blocks of deep neural networks, including attention, normalization, Transformer, BERT, GPT-3, and others, from both biological and algorithmic viewpoints. It also examines generative models like GAN, VAE, normalizing flows, optimal transport, and their underlying geometric principles. By providing a comprehensive understanding of the geometric structure behind deep learning, this book is an invaluable resource for advanced students or researchers seeking to acquire the latest deep learning algorithms and their underlying principles.
Related Products
Related Products
MasterClass in Mathematics Education: International Perspectives on Teaching and Learning
Paul Andrews, Dr Tim Rowland
Artificial Intelligence Engines: A Tutorial Introduction to the Mathematics of Deep Learning
Stone, James V.
Deep Learning with Python
Chollet, Francois
Deep Learning with JavaScript
Cai, Shanqing
Dive into Deep Learning
Zhang, Aston