▶Book Description
In order to build robust deep learning systems, you’ll need to understand everything from how neural networks work to training CNN models. In this book, you’ll discover newly developed deep learning models, methodologies used in the domain, and their implementation based on areas of application.
You’ll start by understanding the building blocks and the math behind neural networks, and then move on to CNNs and their advanced applications in computer vision. You'll also learn to apply the most popular CNN architectures in object detection and image segmentation. Further on, you’ll focus on variational autoencoders and GANs. You’ll then use neural networks to extract sophisticated vector representations of words, before going on to cover various types of recurrent networks, such as LSTM and GRU. You’ll even explore the attention mechanism to process sequential data without the help of recurrent neural networks (RNNs). Later, you’ll use graph neural networks for processing structured data, along with covering meta-learning, which allows you to train neural networks with fewer training samples. Finally, you’ll understand how to apply deep learning to autonomous vehicles.
By the end of this book, you’ll have mastered key deep learning concepts and the different applications of deep learning models in the real world.
▶What You Will Learn
- Cover advanced and state-of-the-art neural network architectures
- Understand the theory and math behind neural networks
- Train DNNs and apply them to modern deep learning problems
- Use CNNs for object detection and image segmentation
- Implement generative adversarial networks (GANs) and variational autoencoders to generate new images
- Solve natural language processing (NLP) tasks, such as machine translation, using sequence-to-sequence models
- Understand DL techniques, such as meta-learning and graph neural networks
▶Key Features
- Get to grips with building faster and more robust deep learning architectures
- Investigate and train convolutional neural network (CNN) models with GPU-accelerated libraries such as TensorFlow and PyTorch
- Apply deep neural networks (DNNs) to computer vision problems, NLP, and GANs
▶Who This Book Is For
This book is for data scientists, deep learning engineers and researchers, and AI developers who want to further their knowledge of deep learning and build innovative and unique deep learning projects. Anyone looking to get to grips with advanced use cases and methodologies adopted in the deep learning domain using real-world examples will also find this book useful. Basic understanding of deep learning concepts and working knowledge of the Python programming language is assumed.
▶What this book covers
- Chapter 1, The Nuts and Bolts of Neural Networks, will briefly introduce what deep learning is and then discuss the mathematical underpinnings of NNs. This chapter will discuss NNs as mathematical models. More specifically, we'll focus on vectors, matrices, and differential calculus. We'll also discuss some gradient descent variations, such as Momentum, Adam, and Adadelta, in depth. We will also discuss how to deal with imbalanced datasets.
- Chapter 2, Understanding Convolutional Networks, will provide a short description of CNNs. We'll discuss CNNs and their applications in CV
- Chapter 3, Advanced Convolutional Networks, will discuss some advanced and widely used NN architectures, including VGG, ResNet, MobileNets, GoogleNet, Inception, Xception, and DenseNets. We'll also implement ResNet and Xception/MobileNets using PyTorch.
- Chapter 4, Object Detection and Image Segmentation, will discuss two important vision tasks: object detection and image segmentation. We'll provide implementations for both of them.
- Chapter 5, Generative Models, will begin the discussion about generative models. In particular, we'll talk about generative adversarial networks and neural style transfer. The particular style transfer will be implemented later.
- Chapter 6, Language Modeling, will introduce word and character-level language models. We'll also talk about word vectors (word2vec, Glove, and fastText) and we'll use Gensim to implement them. We'll also walk through the highly technical and complex process of preparing text data for machine learning applications such as topic modeling and sentiment modeling with the help of the Natural Language ToolKit's (NLTK) text processing techniques.
- Chapter 7, Understanding Recurrent Networks, will discuss the basic recurrent networks, LSTM, and GRU cells. We'll provide a detailed explanation and pure Python implementations for all of the networks.
- Chapter 8, Sequence-to-Sequence Models and Attention, will discuss sequence models and the attention mechanism, including bidirectional LSTMs, and a new architecture called transformer with encoders and decoders.
- Chapter 9, Emerging Neural Network Designs, will discuss graph NNs and NNs with memory, such as Neural Turing Machines (NTM), differentiable neural computers, and MANN.
- Chapter 10, Meta Learning, will discuss meta learning—the way to teach algorithms how to learn. We'll also try to improve upon deep learning algorithms by giving them the ability to learn more information using less training samples.
- Chapter 11, Deep Learning for Autonomous Vehicles, will explore the applications of deep learning in autonomous vehicles. We'll discuss how to use deep networks to help the vehicle make sense of its surrounding environment.
