About kamelconsulting


Why should We Learn Machine Learning?

Many schools around the world including some elementary schools have already started to teach coding classes, the computer science courses for several years now. Coding skills are necessary for us to communicate with computers and there are different ways of writing codes where we do not use complicated mathematics to implement algorithms.

One step further ahead, some schools have even started to teach artificial intelligence in middle schools or in high schools. Machine learning must of course be included in the artificial intelligence courses. How is it possible to learn machine learning even before calculus? It is possible since with some coding experiences the machine learning algorithms can be coded using the available software from the Internet. One can manipulate already built codes to obtain what he wants. Recall, of course, that the necessary computing power is available handy everywhere.

Artificial intelligence is now expected to do the work that people used to do in various fields. They not only replace people in factories or in farms, but they replace some parts of the work in driving, in the military, or even in hospitals. This should imply that artificial intelligence will play various roles in almost every field of the industry. We must get ready for teaching machines so that they can do what we want them to do. That is why we need to learn machine learning courses.

Who are the Readers of this Book?

This book is edited for readers who are new to the machine learning. Readers will get to work with some example problems through which where and how machine learning can be used. To understand the mathematical machine learning algorithm, it is true that one must understand some advanced mathematics based on calculus. In this book, however, we do not intend to describe the mathematics. Instead, we will work on manipulating existing algorithms to do the machine learning example problems.

This book is edited so that readers not specialized in computer science and high school students can learn the process of building and training artificial neural networks by running given computer programs. These programs will guide readers to the solution process of practical problems. A separate lab book is provided to describe only the steps to run the programs. Junior high students may skip reading the main textbook and yet they still can understand the major steps of building artificial neural networks.

Readers who have completed all the example problems in the twelve chapters, should be able to design and formulate neural networks for simple application problems. Those who have completed the exercise problems provided in each chapter with Python programs should be able to solve or at least tackle some application problems.

The Best Way to Start Reading This Book

Starting from Chapter 2, one or two programs are provided in *.exe file format. Readers can download these and start running the programs following steps described in the Lab book. Adult readers and senior high students who are interested in Python programming can also download programs from the “Exercise” subfolder of each Chapter and try to build their own artificial neural networks different from the one in *.exe file.  

Readers who want to create their own neural networks, can run the example programs downloaded, while partially changing the parameters like number of neurons in each layer and changing the input and target data. Readers who are familiar with Python programs will easily be able to apply them to different application problems. All the necessary source code is provided at the same website.

Summary of Artificial Neural Networks Included in the Book

Chapter 1 summarizes different forms of artificial neural networks, Chapter 2 and Chapter 3 describes artificial neural networks for memorizing object names given in the images, Chapter 4 is about an artificial neural network for recognizing English alphabet letters, Chapter 5 is about Hopfield networks for storing and retrieving the whole picture data.

Chapter 6 is about an artificial neural network that classifies color images into four seasons by learning seasonal color images, Chapter 7 includes an artificial neural network that computes the greatest common divisor and the least common multiple for a pair of integers. Chapter 8 describes the steps for creating an artificial neural network that automatically performs a very simple balloon blocking game.

Chapter 9 describes an artificial neural network that performs the addition and subtraction based on the calculation method used by preschool children. Chapter 10 is about an artificial neural network that a learns driving path and drive automatically following the learned path. Chapter 11 starts with descriptions of the convolution and maximum pooling operation on images. Then it describes an example of convolutional neural network that classifies plane diagrams including polygons.

Chapter 12 describes artificial neural networks that memorize the sequences of strokes in writing English cursive alphabets so that they can simulate the act of human writing. Chapter 13 describes an autoencoder neural network applied to reduce the size of pictures.

Program downloads and Contacts

All the computer programs described in this book are available for download from the website below, along with the necessary image files that are required. 

Please contact the author's email below for inquiries or proofreading suggestions regarding the content of the textbook.

Program Downloads: http://www.machinelearningbasic.com

Author contact: byung.s.moon@gmail.com

KamelConsulting, LLC.

CEO. Byung S. Moon | Tel. +82-10-6627-4306 | byung.s.moon@gmail.com ㅣ Biz License 314-86-26086

2F, DICIA ,20 Daedeok-daero 512, Daejeon, S. Korea

2022 Kamel Consulting, LLC