Skip to main content
main-content
Top

About this book

Harness the power of MATLAB for deep-learning challenges. This book provides an introduction to deep learning and using MATLAB's deep-learning toolboxes. You’ll see how these toolboxes provide the complete set of functions needed to implement all aspects of deep learning.
Along the way, you'll learn to model complex systems, including the stock market, natural language, and angles-only orbit determination. You’ll cover dynamics and control, and integrate deep-learning algorithms and approaches using MATLAB. You'll also apply deep learning to aircraft navigation using images.
Finally, you'll carry out classification of ballet pirouettes using an inertial measurement unit to experiment with MATLAB's hardware capabilities.

What You Will LearnExplore deep learning using MATLAB and compare it to algorithmsWrite a deep learning function in MATLAB and train it with examplesUse MATLAB toolboxes related to deep learningImplement tokamak disruption predictionWho This Book Is For
Engineers, data scientists, and students wanting a book rich in examples on deep learning using MATLAB.

Table of Contents

Frontmatter

Chapter 1. What Is Deep Learning?

Abstract
Deep learning is a subset of machine learning which is itself a subset of artificial intelligence and statistics. Artificial intelligence research began shortly after World War II [24]. Early work was based on the knowledge of the structure of the brain, propositional logic, and Turing’s theory of computation. Warren McCulloch and Walter Pitts created a mathematical formulation for neural networks based on threshold logic. This allowed neural network research to split into two approaches: one centered on biological processes in the brain and the other on the application of neural networks to artificial intelligence. It was demonstrated that any function could be implemented through a set of such neurons and that a neural net could learn. In 1948, Norbert Wiener’s book, Cybernetics, was published which described concepts in control, communications, and statistical signal processing. The next major step in neural networks was Donald Hebb’s book in 1949, The Organization of Behavior, connecting connectivity with learning in the brain. His book became a source of learning and adaptive systems. Marvin Minsky and Dean Edmonds built the first neural computer at Harvard in 1950.
Michael Paluszek, Stephanie Thomas

Chapter 2. MATLAB Machine Learning Toolboxes

Abstract
The MathWorks sells several packages for machine learning. Their toolboxes work directly with MATLAB and Simulink. The MathWorks products provide high-quality algorithms for data analysis along with graphics tools to visualize the data. Visualization tools are a critical part of any machine learning system. They can be used for data acquisition, for example, for image recognition or as part of systems for autonomous control of vehicles, or for diagnosis and debugging during development. All of these packages can be integrated with each other and with other MATLAB functions to produce powerful systems for machine learning. The most applicable toolboxes that we will discuss are listed in the following; we will use only the deep learning and the Instrument Control toolboxes in this book.
Michael Paluszek, Stephanie Thomas

Chapter 3. Finding Circles with Deep Learning

Abstract
Finding circles is a classification problem. Given a bunch of geometric shapes, we want the deep learning system to classify a shape as either a circle or something else. This is much simpler than classifying faces or digits. It is a good way to determine how well your classification system works. We will apply a convolutional network to the problem as it is the most appropriate for classifying image data.
Michael Paluszek, Stephanie Thomas

Chapter 4. Classifying Movies

Abstract
Netflix, Hulu, and Amazon Prime all attempt to help you pick movies. In this chapter, we will create a database of movies, with fictional ratings. We will then create a set of viewers. We will then try to predict if a viewer would choose to watch a particular movie. We will use Deep Learning with MATLAB’s pattern recognition network, patternnet. You will see that we can achieve accuracies of up to 100% over our small set of movies. Guessing what a customer would like to buy is something that all manufacturers and retailers want to do as it lets them focus their efforts on products that are of the greatest interest to their customers. As we show in this chapter, deep learning can be a valuable tool.
Michael Paluszek, Stephanie Thomas

Chapter 5. Algorithmic Deep Learning

Abstract
In this chapter, we introduce the Algorithmic Deep Learning Neural Network (ADLNN), a deep learning system that incorporates algorithmic descriptions of the processes as part of the deep learning neural network. The dynamical models provide domain knowledge. These are in the form of differential equations. The outputs of the network are both indications of failures and updates to the parameters of the models. Training can be done using simulations, prior to operations, or through operator interaction during operations.
Michael Paluszek, Stephanie Thomas

Chapter 6. Tokamak Disruption Detection

Abstract
Tokamaks are fusion machines that are under development to produce baseload power. Baseload power is power that is produced 24/7 and provides the base for powering the electric grid. The International Tokamak Experimental Reactor (ITER) is an international project that will produce net power from a Tokamak. Net power means the Tokamak produces more energy than it consumes. Consumption includes heating the plasma, controlling it, and powering all the auxiliary systems needed to maintain the plasma. It will allow researchers to study the physics of the Tokamak which will hopefully lead the way toward operational machines. A Tokamak is shown in Figure 6.1. The inner poloidal field coils act like a transformer to initiate a plasma current. The outer poloidal and toroidal coils maintain the plasma. The plasma current itself produces its own magnetic field and induces currents in the other coils.
Michael Paluszek, Stephanie Thomas

Chapter 7. Classifying a Pirouette

Abstract
A pirouette is a familiar step in ballet. There are many types of pirouettes. We will focus on an en dehors (outside) pirouette from fourth position. The dancer pliés (does a deep knee bend) then straightens her legs producing both an upward force to get on the tip of her pointe shoe and a torque to turn about her axis of revolution.
Michael Paluszek, Stephanie Thomas

Chapter 8. Completing Sentences

Abstract
ABSTRACT
Michael Paluszek, Stephanie Thomas

Chapter 9. Terrain-Based Navigation

Abstract
Prior to the widespread availability of GPS, Loran, and other electronic navigation aids, pilots used visual cues from terrain to navigate. Now everyone uses GPS. We want to return to the good old days of terrain-based navigation. We will design a system that will be able to match terrain with a database. It will then use that information to determine where it is flying.
Michael Paluszek, Stephanie Thomas

Chapter 10. Stock Prediction

Abstract
The goal of a stock prediction algorithm is to recommend a portfolio of stocks that will maximize an investor’s return. The investor has a finite amount of money and wants to create a portfolio to maximize her or his return on investment. The neural network in this chapter will predict the behavior of a stock given its history. This could then be used to select a portfolio of stocks with some idea of the future performance. The stock market model is based on Geometric Brownian Motion. Given that we could do statistical analysis that would allow us to pick stocks. We’ll show that a neural net, which does not have any knowledge of model, can do as well in modeling the stocks.
Michael Paluszek, Stephanie Thomas

Chapter 11. Image Classification

Abstract
Image classification can be done with pretrained networks. MATLAB makes it easy to access and use these networks. This chapter shows you two examples.
Michael Paluszek, Stephanie Thomas

Chapter 12. Orbit Determination

Abstract
Determining orbits from measurements has been done for hundreds of years. The general approach is to take a series of measurements of the object from the ground. This is a set of angles at different times. Given the location on the Earth, and this set of data, one can reconstruct the orbit. Ideal orbits, which make the assumption that the Earth’s gravity is a point at the center of the Earth, are conic sections. Those that stay near the Earth are ellipses. These can be defined as a set of orbital elements. In this chapter, we will design a neural network to find the values for two of the elements. Our model will be simpler than that which astronomers must use. We will assume that all of our orbits are in the Earth’s equatorial plane and that the observer is at the center of the Earth.
Michael Paluszek, Stephanie Thomas

Backmatter

Additional information

Premium Partner

    Image Credits