Deep learning is a central area within artificial intelligence and machine learning. It focuses on training neural networks with many layers to model complex patterns in data. Over the past decade, deep learning has become one of the most influential approaches in AI, with applications in image analysis, natural language processing, generative modeling, and decision-making systems.
A solid conceptual understanding of deep learning is increasingly important for students and professionals working in data science, computer science, and related quantitative fields.
About the book
Understanding Deep Learning by Simon J.D. Prince is a theory-focused text that explains the fundamental ideas behind deep learning models. Rather than emphasizing coding techniques, the book concentrates on the mathematical and conceptual principles that govern how deep neural networks are constructed, trained, and evaluated.
The first part introduces core learning paradigms, including supervised, unsupervised, and reinforcement learning. It then develops the theory of shallow and deep neural networks, loss functions, optimization methods, gradient-based training, and performance evaluation. Later chapters examine specialized architectures for images, text, and graph-structured data, as well as generative models and reinforcement learning.
According to the preface, the early chapters require only introductory linear algebra, calculus, and probability, making them accessible to second-year undergraduates in quantitative disciplines. More advanced chapters assume deeper knowledge of probability and calculus and are intended for advanced students.
The book includes appendices reviewing mathematical prerequisites, so readers do not need to rely on external materials.
What you will learn
Readers will gain a structured understanding of:
- The foundations of supervised, unsupervised, and reinforcement learning
- The structure and theory of shallow and deep neural networks
- Loss functions and maximum likelihood principles
- Optimization techniques such as gradient descent, stochastic gradient descent, momentum, and Adam
- Backpropagation and parameter initialization
- Methods for measuring and improving model performance
- Regularization techniques
- Specialized architectures including convolutional networks, residual networks, transformers, and graph neural networks
- Generative modeling approaches such as GANs, normalizing flows, variational autoencoders, and diffusion models
- Reinforcement learning methods including Q-learning, policy gradients, and actor-critic approaches
- Conceptual questions about generalization, overparameterization, and why deep learning works
- Ethical and social considerations in AI research and practice
The book emphasizes conceptual clarity and theoretical insight, supported by mathematical explanations and illustrations. Appendices cover notation, linear algebra, matrix calculus, and probability theory.
Table of contents
- Preface
- Acknowledgements
- Introduction
- Supervised learning
- Unsupervised learning
- Reinforcement learning
- Ethics
- Structure of book
- Other books
- How to read this book
- Supervised learning
- Supervised learning overview
- Linear regression example
- Summary
- Shallow neural networks
- Neural network example
- Universal approximation theorem
- Multivariate inputs and outputs
- Shallow neural networks: general case
- Terminology
- Summary
- Deep neural networks
- Composing neural networks
- From composing networks to deep networks
- Deep neural networks
- Matrix notation
- Shallow vs. deep neural networks
- Summary
- Loss functions
- Maximum likelihood
- Recipe for constructing loss functions
- Example 1: univariate regression
- Example 2: binary classification
- Example 3: multiclass classification
- Multiple outputs
- Cross-entropy loss
- Summary
- Fitting models
- Gradient descent
- Stochastic gradient descent
- Momentum
- Adam
- Training algorithm hyperparameters
- Summary
- Gradients and initialization
- Problem definitions
- Computing derivatives
- Toy example
- Backpropagation algorithm
- Parameter initialization
- Example training code
- Summary
- Measuring performance
- Training a simple model
- Sources of error
- Reducing error
- Double descent
- Choosing hyperparameters
- Summary
- Regularization
- Explicit regularization
- Implicit regularization
- Heuristics to improve performance
- Summary
- Convolutional networks
- Invariance and equivariance
- Convolutional networks for 1D inputs
- Convolutional networks for 2D inputs
- Downsampling and upsampling
- Applications
- Summary
- Residual networks
- Sequential processing
- Residual connections and residual blocks
- Exploding gradients in residual networks
- Batch normalization
- Common residual architectures
- Why do nets with residual connections perform so well?
- Summary
- Transformers
- Processing text data
- Dot-product self-attention
- Extensions to dot-product self-attention
- Transformers
- Transformers for natural language processing
- Encoder model example: BERT
- Decoder model example: GPT3
- Encoder-decoder model example: machine translation
- Transformers for long sequences
- Transformers for images
- Summary
- Graph neural networks
- What is a graph?
- Graph representation
- Graph neural networks, tasks, and loss functions
- Graph convolutional networks
- Example: graph classification
- Inductive vs. transductive models
- Example: node classification
- Layers for graph convolutional networks
- Edge graphs
- Summary
- Unsupervised learning
- Taxonomy of unsupervised learning models
- What makes a good generative model?
- Quantifying performance
- Summary
- Generative Adversarial Networks
- Discrimination as a signal
- Improving stability
- Progressive growing, minibatch discrimination, and truncation
- Conditional generation
- Image translation
- StyleGAN
- Summary
- Normalizing flows
- 1D example
- General case
- Invertible network layers
- Multi-scale flows
- Applications
- Summary
- Variational autoencoders
- Latent variable models
- Nonlinear latent variable model
- Training
- ELBO properties
- Variational approximation
- The variational autoencoder
- The reparameterization trick
- Applications
- Summary
- Diffusion models
- Overview
- Encoder (forward process)
- Decoder model (reverse process)
- Training
- Reparameterization of loss function
- Implementation
- Summary
- Reinforcement learning
- Markov decision processes, returns, and policies
- Expected return
- Tabular reinforcement learning
- Fitted Q-learning
- Policy gradient methods
- Actor-critic methods
- Offline reinforcement learning
- Summary
- Why does deep learning work?
- The case against deep learning
- Factors that influence fitting performance
- Properties of loss functions
- Factors that determine generalization
- Do we need so many parameters?
- Do networks have to be deep?
- Summary
- Deep learning and ethics
- Value alignment
- Intentional misuse
- Other social, ethical, and professional issues
- Case study
- The value-free ideal of science
- Responsible AI research as a collective action problem
- Ways forward
- Summary
- Notation
- Mathematics
- Functions
- Binomial coefficients
- Vector, matrices, and tensors
- Special types of matrix
- Matrix calculus
- Probability
- Random variables and probability distributions
- Expectation
- Normal probability distribution
- Sampling
- Distances between probability distributions
- Bibliography
- Index
Book details
- Title: Understanding Deep Learning
- Author(s): Simon J.D. Prince
- Main category: Artificial Intelligence
- Subcategory: Deep Learning
- Language: English
- License: Creative Commons CC-BY-NC-ND
More books in: Artificial Intelligence, Deep Learning
Legal notice: This book is shared for educational purposes only. The content is distributed under Creative Commons licenses or with explicit permission from the author. FreeProgrammingBooks may host files that comply with their respective licenses.