Hands-On Neural Networks

Learn how to build and train your first neural network model using Python

Leonardo De Marchi , Laura Mitchell 지음

Packt(GCO Science) 2019년 05월 30일 출간

작품소개

책소개

▶What You Will Learn
- Learn how to train a network by using backpropagation
- Discover how to load and transform images for use in neural networks
- Study how neural networks can be applied to a varied set of applications
- Solve common challenges faced in neural network development
- Understand the transfer learning concept to solve tasks using Keras and Visual Geometry Group (VGG) network
- Get up to speed with advanced and complex deep learning concepts like LSTMs and NLP
- Explore innovative algorithms like GANs and deep reinforcement learning

▶Key Features
- Explore neural network architecture and understand how it functions
- Learn algorithms to solve common problems using back propagation and perceptrons
- Understand how to apply neural networks to applications with the help of useful illustrations

▶Who This Book Is For
If you are interested in artificial intelligence and deep learning and want to further your skills, then this intermediate-level book is for you. Some knowledge of statistics will help you get the most out of this book.

목차

▶TABLE of CONTENTS
1. Getting Started with Supervised Learning
2. Neural Network Fundamentals
3. Convolutional Neural Networks for Image Processing
4. Exploiting Text Embedding
5. Working with RNNs
6. Reusing Neural Networks with Transfer Learning
7. Working with Generative Algorithms
8. Implementing Autoencoders
9. Deep Belief Networks
10. Reinforcement Learning
11. Whats Next?

책 속으로

▶What this book covers
- Chapter 1, Getting Started with Supervised Learning, covers the big picture of AI and, in particular, deep learning. This chapter introduces the main machine learning concepts, from transforming data to evaluating results. These concepts will be useful in the following chapters, where we will focus only on deep learning applications.

- Chapter 2, Neural Network Fundamentals, introduces the building blocks of deep learning and the math behind them. We will also explore concepts such as the perceptron and gradient descent, and the math behind them. We will then see how it's possible to use them to build neural networks with an example, to solve a classification task.

- Chapter 3, Convolutional Neural Networks for Image Processing, covers more complex network architectures for solving domain-specific problems. In particular, we will look at some techniques for solving some computer vision problems. We will also see how a pretrained network can reduce the time needed to create and train a neural network.

- Chapter 4, Exploiting Text Embedding, shows how deep learning can be used for NLP tasks; in particular, how we can use embeddings to process textual data, the theory behind them, and some practical use cases.

- Chapter 5, Working with RNNs, introduces a more sophisticated type of network, RNNs, and the math and the concepts behind them. In particular, we will focus on LSTM and how it can be used to solve an NLP problem.

- Chapter 6, Reusing Neural Networks with Transfer Learning, introduces transfer learning, which is the ability of a model to generalize its learning to different tasks than the one it was trained to solve. We will also look at a concrete example of transfer learning using a pre-trained network to solve our particular problem using Keras and the famous VGG network.

- Chapter 7, Working with Generative Algorithms, introduces one of the most innovative concepts in machine learning in the past decade: GANs. We will see how they work and the math behind them. We will also present an example of how to implement a GAN to generate simple handwritten digits.

- Chapter 8, Implementing Autoencoders, talks about autoencoders, what they are, the math behind them, and which tasks they can solve. In particular, we will look at improvements to the simple autoencoders algorithm and how it's possible to use autoencoders to generate simple handwritten digits with Keras.

- Chapter 9, Deep Belief Networks, talks about Deep Belief Networks (DBNs), what they are, the math behind them, and which tasks they can solve.

- Chapter 10, Reinforcement Learning, introduces RL, starting from the basic concepts, such as the Monte Carlo and Markov chain methods. We will then explain traditional RL methods and how deep learning has improved and revitalized the field.

- Chapter 11, What’s Next?, introduces a quick summary of all the topics that we have covered in the book. We will also provide readers with the details of other titles that could be used as reference materials. Lastly, we will also include the latest advancements that readers can look at in the field of neural networks.

출판사 서평

▶ Preface
Neural networks play a very important role in deep learning and artificial intelligence (AI), with applications in a wide variety of domains, right from medical diagnosis, to financial forecasting, and even machine diagnostics.

Hands-On Neural Networks is designed to guide you through learning about neural networks in a practical way. The book will get you started by giving you a brief introduction to perceptron networks. You will then gain insights into machine learning and also understand what the future of AI could look like. Next, you will study how embeddings can be used to process textual data and the role of long short-term memory networks (LSTMs) in helping you solve common natural language processing (NLP) problems. The later chapters will demonstrate how you can implement advanced concepts including transfer learning, generative adversarial networks (GANs), autoencoders, and reinforcement learning. Finally, you can look forward to further content on the latest advancements in the field of neural networks.

By the end of this book, you will have the skills you need to build, train, and optimize your own neural network model that can be used to provide predictable solutions.

작가정보

저자(글) Leonardo De Marchi

Leonardo De Marchi is an international speaker, author and consultant. He holds a masters in artificial intelligence (AI) and has worked as a data scientist in the sporting world, with clients such as New York Knicks, Manchester United. He now works as a head of data scientist at Badoo, the largest dating site with over 400 million users. He is also the lead instructor at ideai.io, a company specialized in Machine Learning trainings. With Ideai he provides technical and managerial training to large institutions and dynamic startups. He is also a contractor for the European Commission.

저자(글) Laura Mitchell

Laura Mitchell graduated with a degree in mathematics from the University of Edinburgh and, since then, has gained over 12 years' experience in the tech and data science space. She is currently lead data scientist at Badoo, which is the largest online dating site in the world with over 400 million users worldwide. Laura has hands-on experience in the delivery of projects such as NLP, image classification, and recommender systems, from initial conception through to production. She has a passion for learning new technologies and keeping up to date with industry trends.