This book offers a comprehensive introduction to the central ideas that underpin deep learning. It is intended both for newcomers to machine learning and for those already experienced in the field. Covering key concepts relating to contemporary architectures and techniques, this essential book equips readers with a robust foundation for potential future specialization. The field of deep learning is undergoing rapid evolution, and therefore this book focusses on ideas that are likely to endure the test of time.
The book is organized into numerous bite-sized chapters, each exploring a distinct topic, and the narrative follows a linear progression, with each chapter building upon content from its predecessors. This structure is well-suited to teaching a two-semester undergraduate or postgraduate machine learning course, while remaining equally relevant to those engaged in active research or in self-study.
A full understanding of machine learning requires some mathematical background and so the book includes a self-contained introduction to probability theory. However, the focus of the book is on conveying a clear understanding of ideas, with emphasis on the real-world practical value of techniques rather than on abstract theory. Complex concepts are therefore presented from multiple complementary perspectives including textual descriptions, diagrams, mathematical formulae, and pseudo-code.
Chris Bishop is a Technical Fellow at Microsoft and is the Director of Microsoft Research AI4Science. He is a Fellow of Darwin College Cambridge, a Fellow of the Royal Academy of Engineering, and a Fellow of the Royal Society.
Hugh Bishop is an Applied Scientist at Wayve, a deep learning autonomous driving company in London, where he designs and trains deep neural networks. He completed his MPhil in Machine Learning and Machine Intelligence at Cambridge University.
“Chris Bishop wrote a terrific textbook on neural networks in 1995 and has a deep knowledge of the field and its core ideas. His many years of experience in explaining neural networks have made him extremely skillful at presenting complicated ideas in the simplest possible way and it is a delight to see these skills applied to the revolutionary new developments in the field.” — Geoffrey Hinton
“With the recent explosion of deep learning and AI as a research topic, and the quickly growing importance of AI applications, a modern textbook on the topic was badly needed. The “New Bishop” masterfully fills the gap, covering algorithms for supervised and unsupervised learning, modern deep learning architecture families, as well as how to apply all of this to various application areas.” – Yann LeCun
“This excellent and very educational book will bring the reader up to date with the main concepts and advances in deep learning with a solid anchoring in probability. Theseconcepts are powering current industrial AI systems and are likely to form the basis of further advances towards artificial general intelligence.” — Yoshua Bengio
griller –
Warning I just got this new tome, it’s color and I have only started to thumb thru the beginning.
and after reading some more chapters
this book is for someone interested in the basic building blocks of deep learning
like batch norms on differnent levels or variance bias trade off vs double descent overfitting or the nuts of transformers
this is not a cook book
Eric Nichols –
First of all, another great book from Chris Bishop, up to date with Transformers, Graph Neural Networks, and Diffusion Models, etc. that’s all great. Sitting down right now to read the Transformers chapter as a review and to solidify my understanding.
Second, an update: in my earlier review I mentioned some issues with the print quality. I’m pleased to report that this has been fixed. I received a copy from the latest printing and the binding and paper is better. The book as a whole is slightly narrower even though the content is the same: the paper is finer quality than before, and the binding is much nicer so it lays flat more easily. A+ to Springer for improving the quality after the first run.
This book deserves a longer detailed review, but in summary it’s very up-to-date (as of Feb. 2024) and covers all the topics you need to get a deep understanding of deep learning. It’s a great successor to Bishop’s other books on neural networks and machine learning — essential reading for anyone in the field.
Mohamad Issam Sayyaf –
everything was fine.
Oviya Ramachandran Pavanan –
thanks for taking time to write it. I am sure I will benefit greatly.
The book looks and feels like a cheaper international edition – poor paper and print quality. The page margins (left and right) definitely needs rethinking.
Hossein Pishro-Nik –
This is a great book providing a clear blend of theory and practice in the field of deep learning. The book’s approach to deep learning through a probabilistic lens is effective and insightful. It’s an excellent resource for anyone, from beginners to experienced practitioners. The content is well-organized, making complex ideas easy to understand and apply. Overall, a great guide for those looking to understand and use deep learning effectively.
Math Customer –
Great textbook for anyone serious about mastering deep learning. Its exceptional balance of theoretical depth and practical application, combined with high-quality print and well-designed exercises, makes it a standout resource in the field.
TAMU Computer Science –
“This is an excellent book on machine learning and deep learning written by C. Bishop who also wrote a classic book on neural nets in 1995 and a widely used machine learning book in 2006.
The book is extremely well-written, with very deep insights on many topics, including recent developments like large language models and graph neural nets. The math derivations are concise and contain insights. Highly recommended for everyone in AI and ML”.
Shuiwang Ji
Professor and Presidential Impact Fellow
Department of Computer Science and Engineering
Texas A&M University
Claire Edwards –
Highly recommended.
Math Customer –
Received a new copy on Feb 18. Book quality looks fine to me. Loved the content.
Anier Velasco Sotomayor –
About the shipping and manufacturing: it arrived in perfect conditions, great printing quality.About the book itself: it’s just amazing. I had already read some parts of it in the free online version. Great to finally have the hardcover version.
Math Customer –
An excellent book on neural networks, deep learning and their applications to AI and computer vision. Extremely well organized with well-connected chapters which makes it easy to follow and get the BIG picture, while diving deep into the specifics using elaborate math language and yet not lacking the intuitive approach. The book can be used either as a one stop-reference for subject matter experts, or simply turned into a self-learning or university level textbook. This book also enjoyed the exceptional and unique style of Prof. Christopher M Bishop , who has been a researcher, a public speaker, and a distinguished leader in the field of NN, DL, ML and AI for decades.Dr. Cazhaow QazazVP of Advanced Analytics and AIKnowledge Square LLC
Math Customer –
Hard cover but I expected the paper quality to be smooth like Elements of Statistical Learning…
mkj –
I have many books on Deep Learning and this is the first that presents the subject in a coherent, well thought out manner.
Each topic is introduced carefully, explained well with relevant pointers to other fields/work. The underlying statistics are clearly presented in a logical manner. The connections between many of the techniques used are clearly exposed.
It’s a delight to read and I would recommend it wholeheartedly.
C. Jaensch –
A high-quality textbook on deep learning that covers both the fundamentals – including an introduction to probability theory – as well as recent developments, such as Transformer models, autoencoders, or diffusion models. I especially love the targeted use of color to accentuate important points or elements in diagrams.
One can tell that the main author has decades of experience in the field and spent years compiling and writing this book.
Readers should be aware that this is a university-level book that requires at least a calculus and linear algebra background to be able to follow the explanations.
Longo Giuseppe –
I am a professor of Machine learning. My interest in the field arose from reading the first book by Bishop which, in my opinion, was the best book ever written on the topic: clear, exhaustive and inspirational. Unfortunately over the years it became outdated. I was therefore thrilled when I saw this new book and I bought it immediately and read eagerly. It is a masterpiece, even better and more seminal than the previous one. I adopted it for my course on Machine learning. In my opinion it is a must have in the library of every person working in the field: theoretician or practitioner. Thanks Prof. Bishop.
Math Kunde –
Very nice book. Awesome content.
Andrea Pranzo –
High quality print and content
Insoo Chung –
I love the printed version of the book and I like how readable this book is. English is not my mother tongue and I learned math in my language (i.e. not English) so CS textbooks are very hard to comprehend for me. But this is such a breeze to read. Also, the print/paper quality is one of the best I’ve seen on books in the US market.
+ I noticed that there are some binding quality issues for other reviewers – mine came in great condition.
Jan –
Great introduction to the realm of ML and very very well organized survey of fundamentals. I really do appreciate how perfectly this book is written and effort put into wording. Everything fits properly and is systematically built like back in my university courses. Exactly what I was looking for: textbook and reference at the same time. Not overhelmingly rigorous and definitely not too vague and sloppy. Affordable price tag is also a huge bonus. Highly recomended!
Daniel Deychakiwsky –
I highly recommend this book to any practicing machine learning engineers. It’s a great technical reference for applied deep learning.
Saurabh –
It covers required mathematical details for deep learning.
Eduardo Hiroshi Nakamura –
Muito bom!
Vamsi –
As far as deep learning textbooks go, this is as comprehensive and well-written as you can hope for. The writing is tight — each sentence of each section is easy to understand and succinct. Highly recommend!
AT –
This book features high-quality glossy pages and color printing, enhancing readability. It provides intuitive explanations of key concepts and excellent pedagogy on deep learning, covering both basic principles and recent advancements.
Panagiotis N. Zarros –
I bought this books to acquaint myself on deep learning, but the 2 chapters “Convolutional networks” and “Transformers” are written badly and did not get anything out of this other than a general architecture. How the information is passed and used from one layer to the other is completely missed. Besides those 2 chapters on deep learning, the rest of the book is on topics related to machine learning is easy to follow (if you have some decent background on matrix analysis and statistics).
Ritter –
I’m feeling lucky.
This is incredibly, unspeakably, impossibly good book.
I’ve seen the book at the CVPR conference, and hesitated a bit before buying it. $90 after all.
But I can’t be happier with my purchase. It’s simply the best book on Deep Learning – and on Machine Learning – I’ve seen in 20 years. It gives a very accessible and intuitive introduction in many topics – and, at the same time, preserves the mathematical rigor. For example, the chapter on transformers is the very, very best description of a topic I’ve seen so far.
The book also contains all the info on the ML basics – from multivariate distributions to EM algorithm. Again, it is written very candidly – and, at the same time, preserves theoretical foundations. It is probably slightly less “mathy” than the “Pattern Recognition and Machine Learning” book by Bishop – which is a previous iteration of this text. But much, much more accessible as well. It explains “narrow” topics in plain English. You may read it in this textbook, and then go back to “Pattern Recognition” if you want to get a slightly deeper math for the same topic.
Also the book is very recent and covers the hottest and most recent topics, like Diffusion Models.
Saved me lots, lots of time and effort – instead of digging through a heap of research papers I can nos just read a chapter from this book.
Many thanks to Chris Bishop for writing it!
Nirmal –
Deep Learning: Foundations and Concepts is the best book to learn fundamentals of Neural Networks and Deep Learning.
Ruby Lozano de Angulo –
I was impressed with the way this book was written. It’s an excellent introduction to anyone wanting to learn about DL practices in a general level. It also has a hard-cover and excelent print quality. The authors only had a slip along the way for transformers, but I hope they can improve upon it on their next publication.
Pragya Kale –
Best of the best
Yakub kadri İlhan –
İçerik muhteşem, onu zaten herkes biliyor. Cilt kalın ve dayanıklı, yapraklar kalın ve dijital baskı, hunharca kullanılmadığı müddetçe bir sorun çıkmaz.
Nguyen Khiem –
What are the contents or Chapters of thick book
Math Digital –
Foundations: Probability theory, linear algebra
Neural Networks: Perceptrons, multi-layer perceptrons, backpropagation
Deep Learning Architectures: Convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), transformers
Learning Algorithms: Stochastic gradient descent, regularization techniques
Practical Applications: Computer vision, natural language processing, speech recognition