Book Image

Deep Learning for Beginners

By : Dr. Pablo Rivas

Book Image

Deep Learning for Beginners

By: Dr. Pablo Rivas

Overview of this book

With information on the web exponentially increasing, it has become more difficult than ever to navigate through everything to find reliable content that will help you get started with deep learning. This book is designed to help you if you're a beginner looking to work on deep learning and build deep learning models from scratch, and you already have the basic mathematical and programming knowledge required to get started. The book begins with a basic overview of machine learning, guiding you through setting up popular Python frameworks. You will also understand how to prepare data by cleaning and preprocessing it for deep learning, and gradually go on to explore neural networks. A dedicated section will give you insights into the working of neural networks by helping you get hands-on with training single and multiple layers of neurons. Later, you will cover popular neural network architectures such as CNNs, RNNs, AEs, VAEs, and GANs with the help of simple examples, and learn how to build models from scratch. At the end of each chapter, you will find a question and answer section to help you test what you've learned through the course of the book. By the end of this book, you'll be well-versed with deep learning concepts and have the knowledge you need to use specific algorithms with various tools for different tasks.

Preface

Who this book is for

What this book covers

To get the most out of this book

Section 1: Getting Up to Speed

Section 1: Getting Up to Speed

Free Chapter

Introduction to Machine Learning

Introduction to Machine Learning

Diving into the ML ecosystem

Training ML algorithms from data

Introducing deep learning

Why is deep learning important today?

Questions and answers

Setup and Introduction to Deep Learning Frameworks

Setup and Introduction to Deep Learning Frameworks

Introduction to Colaboratory

Introduction and setup of TensorFlow

Introduction and setup of Keras

Introduction to PyTorch

Introduction to Dopamine

Other deep learning libraries

Questions and answers

Preparing Data

Binary data and binary classification

Categorical data and multiple classes

Real-valued data and univariate regression

Altering the distribution of data

Data augmentation

Data dimensionality reduction

Ethical implications of manipulating data

Questions and answers

Learning from Data

Learning from Data

Learning for a purpose

Measuring success and error

Identifying overfitting and generalization

The art behind learning

Ethical implications of training deep learning algorithms

Questions and answers

Training a Single Neuron

Training a Single Neuron

The perceptron model

The perceptron learning algorithm

A perceptron over non-linearly separable data

Questions and answers

Training Multiple Layers of Neurons

Training Multiple Layers of Neurons

Minimizing the error

Finding the best hyperparameters

Questions and answers

Section 2: Unsupervised Deep Learning

Section 2: Unsupervised Deep Learning

Autoencoders

Introduction to unsupervised learning

Encoding and decoding layers

Applications in dimensionality reduction and visualization

Ethical implications of unsupervised learning

Questions and answers

Deep Autoencoders

Deep Autoencoders

Introducing deep belief networks

Making deep autoencoders

Exploring latent spaces with deep autoencoders

Questions and answers

Variational Autoencoders

Variational Autoencoders

Introducing deep generative models

Examining the VAE model

Comparing a deep and shallow VAE on MNIST

Thinking about the ethical implications of generative models

Questions and answers

Restricted Boltzmann Machines

Restricted Boltzmann Machines

Introduction to RBMs

Learning data representations with RBMs

Comparing RBMs and AEs

Questions and answers

Section 3: Supervised Deep Learning

Section 3: Supervised Deep Learning

Deep and Wide Neural Networks

Deep and Wide Neural Networks

Wide neural networks

Dense deep neural networks

Sparse deep neural networks

Hyperparameter optimization

Questions and answers

Convolutional Neural Networks

Convolutional Neural Networks

Introduction to convolutional neural networks

Convolution in n-dimensions

Convolutional layers

Pooling strategies

Convolutional neural network for CIFAR-10

Questions and answers

Recurrent Neural Networks

Recurrent Neural Networks

Introduction to recurrent neural networks

Long short-term memory models

Sequence-to-vector models

Vector-to-sequence models

Sequence-to-sequence models

Ethical implications

Questions and answers

Generative Adversarial Networks

Generative Adversarial Networks

Introducing adversarial learning

Comparing GANs and VAEs

Thinking about the ethical implications of GANs

Questions and answers

Final Remarks on the Future of Deep Learning

Final Remarks on the Future of Deep Learning

Looking for advanced topics in deep learning

Learning with more resources from Packt

Other Books You May Enjoy

Other Books You May Enjoy

Leave a review - let other readers know what you think

Customer Reviews

5 star

0

4 star

0

3 star

0

2 star

0

1 star

0

Training ML algorithms from data

A typical preprocessed dataset is formally defined as follows:

Where y is the desired output corresponding to the input vector x. So, the motivation of ML is to use the data to find linear and non-linear transformations over x using highly complex tensor (vector) multiplications and additions, or to simply find ways to measure similarities or distances among data points, with the ultimate purpose of predicting y given x.

A common way of thinking about this is that we want to approximate some unknown function over x:

Where w is an unknown vector that facilitates the transformation of x along with b. This formulation is very basic, linear, and is simply an illustration of what a simple learning model would look like. In this simple case, the ML algorithms revolve around finding the best w and b that yields the closest (if not perfect) approximation to y, the desired output. Very simple algorithms such as the perceptron (Rosenblatt, F. 1958) try different...