Deep Learning (Adaptive Computation and Machine Learning series)

This book is possibly currently unique in its coverage of the latest ideas in the field of deep learning -- and it is a very convenient and good survey of fundamental concepts (linear algebra, optimization, performance metrics, activation function types), different network types (multi-layer perceptron, convolutional neural networks, and recurrent neural networks), practical considerations (data set, training and validation, implementation), and applications (comments on existing real-world/commercial uses). The final 235 pages of the content portion of the book is dedicated to topics in "Deep Learning Research", and these topics are truly at the current frontier. Another reviewer said that one could gain the same knowledge of cutting-edge research by reading all of the latest papers (from academia and industry), but the "research" section of this book offers the following: Selection of the most notable research by the very experienced authors of the book, and collection of similar research in to a broader discussion of themes, and the additional insights. The book covers very advanced and new ideas currently being explored, and it is very nice to be able to have a consistent and coherent presentation of all of those ideas. However, the book is also packed with valuable observations and pointers about more basic aspects of deep learning implementations and practices -- and such commentary is in depth and includes substantial analysis and mathematical derivation (in an intuitive presentation that often includes graphs illustrating the phenomenon). As someone with an intermediate level of knowledge and experience of neural networks, I am really grateful for this book, because seems like the ideal resource for learning cutting-edge ideas and practices, with context. The book has excellent scope and depth, and I am confident that anyone with a solid background in linear algebra, calculus, statistics, and general machine learning, and basic neural networks (multi-layer perceptrons) will find this book to be very exciting and perhaps unique in its ability to take the reader to the next level and a new frontier. I was personally excited to learn about the idea of representing the dependencies of intermediate quantities by directed graphs, and how this can be used to perform calculations for recurrent neural networks efficiently. And I think the long chapter on recurrent neural networks is very helpful. Having said all of this, I think only people with significant working knowledge and experience with neural networks and mathematics -- people whose academic or professional focus has been neural networks for at least a year or two -- would benefit from this book. This book answers a lot of the deeper questions that one is likely to have while developing a solid understanding of the fundamentals, and that's one of the book's tremendous values, but this book assumes an understanding of the fundamentals (but does briskly cover the basics). I think this book is a perfect follow-up book for the excellent book "Neural Network Design (2nd edition)" by Hagan, Demuth, Beale, and de Jesus, and I highly recommend the latter for gaining the solid background needed to have a thrilling experience with the "Deep Learning" book. In summary, I am very glad this "Deep Learning" book was written, and I think the "Deep Learning" book will be a great benefit to a lot of people, and to the evolution of the field.

Fantastic deep-learning book! The logic is very easy to follow, but the content is very thorough when it comes to explaining the theories behind it, making it perfect for beginners as well as math and CS students. The best DL/ML book I have ever seen!!

This is not a coding book. I see a lot of negative reviews around the expectation that this book would teach the reader how to quickly build machine learning systems and write code. This book is not for that audience. If you just want to build applications, don't worry about how deep learning works. It's akin to needing to understand how an engine works just to drive a car. If you are looking for a coding resource, try: https://www.amazon.com/Hands-Machine-Learning-Scikit-Learn-TensorFlow/dp/1492032646/ref=sr_1_4?keywords=machine+learning+tensorflow&qid=1579608765&sr=8-4 . And even with that book, the material still goes far beyond what you need - use it as a light reference. I bought this book as an aspiring machine learning researcher, and towards that end, it is the best resource available in print (still true as of 2020). For instance: The first 5 chapters are timeless. These are things that were mostly established 20 or 30 years ago and beyond and are mostly STEM fundamentals at this point. There are whole textbooks dedicated to each of those chapters, but the authors provide a quick refresher and overview of probably 80% of what you'll encounter in deep learning. If you haven't previously learned each of these subtopics, you'll probably want to study them individually since they are the key to innovating (linear algebra, probability & stats, numerical computation, machine learning fundamentals). Chapters 6 thru 9 are the foundation of deep learning. We're about 12 years into seeing rapid change in the deep learning space, yet all of these principles and techniques still hold (many recent innovations are still relying on Convolutional models in 2020, which is the most layered/complex topics in those chapters). Therefore, I'd wager that these chapters are also fairly stable knowledge that is worth internalizing if you want to be deeply involved in the future of machine learning. Chapters after 9 are mostly experimental topics, and many of them are already the wrong strategies for optimal results. But there are interesting ideas in here that you'll often encounter in the wild, so it's good exposure to various topics. But probably not worth much of your time. And lastly, there is good history in here from people who know the space intimately. It's a good way to piece together the developments and learn the lexicon of deep learning so you can have intelligent conversation with experts.

Deep Learning is an advanced book with great explanations and details. There is a heavy math focus with the book's beginning chapters detailing the necessary linear algebra and probability that one will need to understand deep learning. I liked that the author's chose to cover only the parts of these subjects which are relevant to deep learning. There are many interesting philosophical sections in the book as well. Just about when I was feeling overwhelmed with the complexity of the mathematics the authors take a step back and cover the foundations of deep learning such as borrowing concepts from human learning. There was an interesting dicussion about the early studies done on the vision of cat's and monkey's in the 1970s. The text covers the entire history of deep learning and the bibliography is hundreds of sources. It is clear this is the most comprehensive text available about deep learning. For anybody interested in this topic this book is a mandatory read. There are sections about machine learning as well, which makes sense because deep learning is a subset of machine learning. These sections focused on the machine learning concepts which are most relevant to deep learning. The book was well organized and divided into three parts which cover mathematics related to deep learning, typical deep learning techniques, and then more experiment learning techniques. Often the author's state when a technique works well or when it does not, and which types of data works best for the technique. Just a warning, the math in this book is highly complex. It requires a lot of work to go through this book, but the effort will be well rewarded.

A lot of reviews bashing the book, but it's easy to criticize what the weaknesses are in any long piece of work. But what are the better alternatives? As far as I know it's the best overview on the subject currently out there. If you read this, Reinforcement Learning by sutton and barto, and watch some free lectures on YT from any decent university, you will know a lot about machine learning. As for ML in practice, most people are just copy pasting code from stackoverflow/random blogs/tutorials and using ML libraries, they have no idea how any of this works and can't be bothered to learn.

The best book that you can buy to learn

This book, in every sense of the word, is rushed. I think the authors wanted to establish themselves as leaders of this young-ish field, but does so by sacrificing quality. It also shows that Deep Learning theory has been there for a long time, known by another name called Neural Networks. The interesting algorithms are of MLP, Back Propagation and the classical neural networks. The optimization methods such as Adam are the ones that are new and interesting, and the only ones worthy of in this book. So, essentially, what you get from this book is use A for X, B for Y and C for Z type of dry, un-intuitive, badly written waste of paper. As for the structure of the book, it's like an example of how not to structure a book. It has some linear algebra, probability at the start (not good enough, and confuses more people and wastes paper). Goes on to prove other algorithms such as PCA (yeah, ok!). Then, talks about how this architecture works for this and that architecture. So, yeah, if you really want to try out deep learning, don't buy this book. Set up Tensorflow/pytorch/ other library, run the tutorials, find an architecture for the problem you are interested in and start tweaking that. You will have far more fun and would have saved your money. The praise that this book gets is beyond me. Did Musk even read this book? I doubt it.

This book has by far surpassed my expectations! I have purchased many machine learning and deep neural network books in the past, but nothing has ever come close to this book! First of all, it is written by the fathers of Deep Learning, and is therefore an authority. Secondly, the book is broken into three parts: 1. A math overview and refresher. 2. Deep Learning applications and 3. Research in Deep Learning. I can't help but go through this book from front to back. It is a smooth read, and every sentence written is meaningful. These guys know their stuff! And after you read this book, YOU WILL ALSO know your stuff! If you feel daunted by the price, just remember, you get what you pay for! I'd say they could easily charge about $300+ for this book, but they are doing everyone a very kind favor by ONLY charging this reasonable amount. You get A LOT of bang for your buck with this purchase. I hesitated at first about buying this book because of the price, but I am soooooo happy that I did! Worth every penny! Look no further, get this book and start your Deep Learning journey!!

Libro excepcional sobre deep learning, considerado una de las principales referencias en la materia. Es increíblemente completo y profundo, cubre desde los fundamentos hasta conceptos avanzados, con un enfoque muy académico. Sin embargo, no es un libro sencillo: requiere una base sólida en matemáticas y machine learning para poder aprovecharlo bien. Quizás resulte más útil para investigadores, doctorandos o quienes quieran profundizar a nivel teórico que para principiantes. Aun así, es una obra imprescindible en cualquier biblioteca de inteligencia artificial.

Great book for aspiring deep learning enthusiasts

helpful for anymore who wants an introductory (and broad) background to the field

Item arrived damage, with badly printed pages. I tried to contact the seller, but messages cannot be delivered to it.

Las bases del aprendizaje profundo están en este libro, un buen conocimiento en matemáticas avanzadas es crucial