| Neural Networks and Deep Learning A Textbook 목차
|Table of Contents
1 An Introduction to Neural Networks.- 2 Machine Learning with Shallow Neural Networks.- 3 Training Deep Neural Networks.- 4 Teaching Deep Learners to Generalize.- 5 Radical Basis Function Networks.- 6 Restricted Boltzmann Machines.- 7 Recurrent Neural Networks.- 8 Convol
utional Neural Networks.- 9 Deep Reinforcement Learning.- 10 Advanced Topics in Deep Learning.
5.0 out of 5 starsThe best book in the academic genre
October 4, 2018
This is a fantastic book from the academic perspective, and has
quite a bit for practitioners too in terms of conceptual understanding.
Considering the fact that it is a mathematically intensive book,
it is relatively easy to understand. Not an implementation book, but great for
deeply understanding concepts. The book has managed to provide
discussions of the architecture of lots of real-world applications of neural
networks in text, images, among others, which is good for
practitioners. Certainly, hands down better than the Goodfellow book,
the only other directly comparable book out there in terms of style
and material covered.
5.0 out of 5 starsDefinitely a Must Have - If you are interested in Neural Networks and Deep Learning
November 1, 2018
I have familiarity with data mining concepts and general machine learning. I am a practitioner of Machine Learning and am very interested in applying these models to real world problems. The purpose of buying this book was two fold: 1. I wanted to get an understanding of deep learning - how neural networks work and how they can be put to use and 2. How do neural networks compare in relation to other conventional machine learning models? How are they related and where is their place in the field of machine learning?
From either point of view, I feel that every penny I spent on buying the book is worth more than its weight in gold. This book starts with a fairly detailed introduction into simple neural networks. The early chapters establish crucial and very useful connections between conventional machine learning methods and how neural networks can be built to mimic them. Ample examples and details are given to walk the user through intricate scenarios. Example, there is a whole section which unboxes gradient descent and explains the math behind it. There are several places in the book where connections are drawn between neural networks and how they simulate linear regression, logistic regression and SVMs. Several variants and differences are also explained in great detail. Once these are established, early development in neural networks are addressed - Radial Basis Functions and Restricted Boltzmann Machines are discussed in depth. After setting the fundamentals, the author goes on to address topics in deep learning - starting with RNNs, CNNs, Deep Reinforcement Learning and more advanced topics like GANs.
The book also provides and cites ample references which inform the user about the historical progress and development of the field. The references have been compiled with great care and so are the diagrams. Very detailed explanations are provided to connect practicality of the methods. For instance, for activation functions, several examples are provided based on what functions are used in practice and how the choice impacts the complexity of models and what conventional ML models they map to.
A more detailed review will follow as I progress more through the book but for starters, this is a great book to buy - be it for reference, or teaching a course or for getting to know the field. If you have experience in ML, you will definitely benefit from the insightful connections of neural networks with conventional ML methods. For teaching, the accompanying web page has a wealth of resources in the form of slides, Image sources for pictures in the book to compose your own slides and other files accompanying the book. Definite buy to have in your shelf if you are interested in Deep Learning.
5.0 out of 5 starsCan't recommend it enough
December 12, 2018
I've truly enjoyed this book. I don't have a mathematical background, so some machine learning techniques can be difficult to understand without a lot of serious effort. I haven't had that problem with this book. It explains how various neural networks work at a conceptual level, which is a must-have for anyone considering doing serious work in the field. Even though it's math intensive, I found it very easy to understand and the figures were incredibly helpful in piecing everything together. It is also very comprehensive. For the past year, I have been doing survey research in the field and this book is thorough: it goes in detail on every major model and advancement.
Just keep in mind that this is not a technical how-to, it focuses mainly on conceptual understanding.
It is not easy to simplify a complex and difficult field in such a well-thought out way. I'd recommend it to researchers, students and everyone else interested in deep learning.
5.0 out of 5 starsGreat book!
November 16, 2018
The "Neural Networks and Deep Learning" book is an excellent work.
The material which is rather difficult, is explained well and becomes understandable
(even to a not clever reader, concerning me!).
The overall quality of the book is at the level of the other classical "Deep Learning" book
of Goodfellow, both books are outstanding and can help to provide
their own views at the exploding domain of deep learning.
In my opinion the reader can better benefit from studying both books together,
they provide complementary insights to the subtle and intricate mathematics and
5.0 out of 5 starsComplex material is well simplified
October 10, 2018
Finally, we have a book that combines intuition and mathematics to
describe the analytical and methodological aspects of deep learning.
The writing style makes it easy to follow complex material.
This type of approach is needed for true mastery of the subject.
In this respect, it is probably a great complement to implementation-style books,
because academic books have a focus on fundamentals rather than implementation
frameworks. The book also provides intuition on how neural networks can be used
in many real-world applications. In that sense, I do not think that
the utility of this book is restricted to academia.
5.0 out of 5 starsAn academic oriented textbook for deeper understanding of DL
October 29, 2018
This book is in my opinion a welcome addition to deep learning resources. There are a few books on the practical know-how and software implementation for neural networks. This book provides a deeper understanding of the concepts and the algorithmic methods and requires a background in multi-variate calculus and linear algebra. Useful exercises enhance its pedagogical value. Chapters also list recommended video lectures and software resources for the practitioner. This book should be a useful GOTO to also DL practitioners seeking deeper understanding to be successful in their applications.
5.0 out of 5 starsA truly good textbook
September 22, 2018
This is a truly outstanding textbook on deep learning, with broad coverage, deep analysis as well as well-thought exercise. Good for researcher, graduate students as well as practitioners. highly recommended. The author is well-established and a top person in data mining and machine learning. He also has written many other great textbooks over years.
5.0 out of 5 starsexcellent
November 5, 2018
An Excellent book for students, researchers and engineers at all levels.
This book tells you the most important background, theories, models, methods and practices in deep learning. Unlike many serious books that are hard to understand, it is really enjoyable to read this book that clearly explains how the whole stuff works together from the very begining to the very end in concise and most accurate language.
I have been working in this area for more than one decade from the era when we simply called "deep learning" "neural networks," and well followed it evolving into deep learning and AI. But I still learn a lot and refresh my idea when I read many chapters in this book.
Definitly will recommend it to all my friends and colleagues.
| 도서 상세설명
|This book covers the theory and algorithms of deep learning and it provides detailed discussions of the relationships of neural networks with traditional machine learning algorithms.
The mathematical aspects are concretely presented without losing accessibility.
The book is written in a textbook style, and it includes exercises, a solution manual, and instructor slides. The depth and breadth of coverage are unique to the book
About this Textbook
This book covers both classical and modern models in deep learning. The primary focus is on the theory and algorithms of deep learning. The theory and algorithms of neural networks are particularly important for understanding important concepts, so that one can understand the important design concepts of neural architectures in different applications. Why do neural networks work? When do they work better than off-the-shelf machine-learning models? When is depth useful? Why is training neural networks so hard? What are the pitfalls? The book is also rich in discussing different applications in order to give the practitioner a flavor of how neural architectures are designed for different types of problems. Applications associated with many different areas like recommender systems, machine translation, image captioning, image classification, reinforcement-learning based gaming, and text analytics are covered. The chapters of this book span three categories:
The basics of neural networks: Many traditional machine learning models can be understood as special cases of neural networks. An emphasis is placed in the first two chapters on understanding the relationship between traditional machine learning and neural networks. Support vector machines, linear/logistic regression, singular value decomposition, matrix factorization, and recommender systems are shown to be special cases of neural networks. These methods are studied together with recent feature engineering methods like word2vec.
Fundamentals of neural networks: A detailed discussion of training and regularization is provided in Chapters 3 and 4. Chapters 5 and 6 present radial-basis function (RBF) networks and restricted Boltzmann machines.
Advanced topics in neural networks: Chapters 7 and 8 discuss recurrent neural networks and convolutional neural networks. Several advanced topics like deep reinforcement learning, neural Turing machines, Kohonen self-organizing maps, and generative adversarial networks are introduced in Chapters 9 and 10.
The book is written for graduate students, researchers, and practitioners. Numerous exercises are available along with a solution manual to aid in classroom teaching. Where possible, an application-centric view is highlighted in order to provide an understanding of the practical uses of each class of techniques.
Resources and teaching slides at author website
| 교육용 보조자료