Artificial Intelligence Simplified: Understanding Basic Concepts

Artificial Intelligence Simplified

Understanding Basic Concepts

Authors:

Binto George

Gail Carmichael

Editors:

Susan S. Mathai

Andrew Carmichael

Copyright ©2015 CSTrends LLP.

All rights reserved.

Publisher's Website:

www.cstrends.com

ISBN:1-944708-02-2

ISBN-13: 978-1-944708-02-3

Library of Congress Control Number: 2015920904

HOW TO USE THIS BOOK

Artificial Intelligence (AI) will impact our lives in more ways than we can imagine. This book is for anyone wanting to learn AI concepts without an in-depth prior knowledge in the field. You should be able to use this book if you have some basic background in data structures and algorithms.

The book introduces key AI concepts in an easy-to-read format with examples and illustrations. Our emphasis is to keep our explanations as simple as possible. We feel that a complex, overly mathematical textbook does not always serve the purpose of conveying basic AI concepts to most people.

If you are a professional and wish to learn AI without complexities, this book is for you. If you are a robotics enthusiast wanting to understand the broader aspects of AI, you may find this book useful. If you are taking a basic AI course and find the traditional AI textbooks intimidating, you may use this as a bridge book, or as an introductory textbook.

DISCLAIMER

Authors and publisher make no representations or warranties on accuracy or completeness with respect to the information provided in this book. No implied warranties of fitness or merchantability are made for specific purposes. No one is authorized to offer any expressed or implied warranties on our behalf. Always seek appropriate professional advice before using any information provided in this book. Neither authors nor publisher will be liable for any losses or damages including without limitation, indirect or consequential loss or damage.

Product and company names mentioned in this book may be trademarks or registered trademarks of their respective holders. Their use or appearance does not imply any affiliation with or endorsement by them.

Table of Contents

1. Introduction

1.1. Organization

1.2. The Operating Room Scheduling Problem

1.3. Generate and Test

2. Scheduling with Search Methods

2.1. Blind Search Methods

2.2. Heuristic Search Methods

2.2.1 Hill Climbing

2.2.2 Best First Search

2.3. Best Path Methods

3. Accommodating Surprises with Planning Techniques

3.1. Forward Planning

3.2. Backward Planning

3.3. Partial-Order Planning

3.4. Planning Under Uncertainty

4. Evolving Schedules with Genetic Algorithms

4.1. Genetic Programming

5. Learning from Experience With Neural Networks

5.1. Multi-layer neural networks

6. Expert Systems for Diagnosis

6.1. Expert System Types

6.1.1 Forward chaining

6.1.2 Backward chaining

6.1.3 Hybrid chaining

6.1.4 Deduction and reaction systems

6.2. Fuzzy Expert Systems

7. Handling Competing Goals With Game Trees

8. Communicating With Natural Language

8.1. Natural Language Understanding

9. Identifying Intelligence

9.1.1 Super Intelligence

10. Conclusions and Where to Go From Here

10.1. AI and Other Disciplines

10.2. The Future of Artificial Intelligence

Appendix A: Search Methods

A.1. Depth First Search (DFS)

A.2. Breadth First Search (BFS)

A.3. Simple Hill Climbing

A.4. Steepest Ascent Hill Climbing

A.5. Best First Search

A.6. A*

Appendix B: Neural Network Learning More Complex Logic

Appendix C: Fuzzy Expert System

Bibliography

INDEX

1. INTRODUCTION

Objectives

Introduce AI

Explore applications of AI

Discuss the organization of the book

Describe the operating room scheduling problem

Introduce the generate-and-test problem solving method

If you have ever checked the weather, snapped a picture with your smart phone, or even just searched for information online, you have used Artificial Intelligence (AI). Modern applications of AI range from computer games to self-driving cars. Financial institutions use AI for fraud monitoring, investment decisions support, credit risk assessment, data mining of customer behaviors, and economic forecasting. The military uses AI for target discrimination, missile defense shields, and robot steering.

In the coming years, applications of AI are likely to touch every walk of life. We created this book to help you understand some of the major concepts of AI to help you succeed in such an environment.

You probably won’t become an AI expert, or even know every detail of the concepts we present. But you will get a sense of some of the central ideas in AI through a context you can relate to.

The original objective of AI was to create systems that perceive, think and act like humans (Winston, 1992). Years of research have shown us that this is a tougher proposition than originally anticipated. While a few researchers continue their work towards achieving this objective, most others work instead toward the objective of developing systems that perform as good as or better than humans in a focused area such as chess playing or disease diagnosis.

Researchers have tried many different methods to achieve AI. Some believe that symbolic systems hold the key to the AI. A few believe heuristic searchers are the way to go. Others think that expert systems can capture human intelligence. Some focus on recreating the human brain by simulating neurons in the brain. In this book, we explain some of these methods in the context of realistic applications.

1.1. Organization

In this book, we tackle several different problem areas that can be addressed with AI techniques. For example, AI can help job scheduling in both the manufacturing and service industries by automatically generating high quality schedules while considering various parameters and meeting potentially conflicting objectives. AI can help find driving directions from one location to another through a network of roads with real-time updates based on dynamic events such as detours, constructions and accidents. If you are traveling to a number of cities and you want to return after visiting all cities exactly once, which route is the best for you? Questions like that can be solved using the search methods described in Chapter 2.

Planning is the process of deciding the sequence of actions to be performed for reaching one of the goal states starting from the current state. Planning helps us deal with surprise events that occur during otherwise normal course of operations. We can apply planning techniques to such domains as logistics, military campaigns, space exploration, and operating procedure synthesis. Chapter 3 explains planning.

There is an interesting AI method used for gaming, cryptographic code breaking, freight routing, data packet routing, market prediction, hardware design, signal filtering and signal processing. Genetic algorithms have origins in evolutionary biology, in which organisms evolve and adapt to thrive in environmental conditions. Learn more about genetic algorithms and evolutionary computing in Chapter 4.

Learning from experience is a sign of intelligence. Neural networks simulate the learning capacity of biological neurons in our brain. Neural networks can be effectively used for gesture recognition, speech recognition, handwriting recognition, fraud detection, cancer cell detection and petroleum exploration, and much more. In particular, big data systems handle huge volume and variety of data moved at extremely high speeds, where conventional data processing methods are not sufficient. Neural networks can predict patterns or devise processing strategies to sufficiently deal with big data. See Chapter 5 for more information on neural networks.

Expert systems can assist or even replace human experts with specialized knowledge. Expert systems for disease diagnosis include MYCIN (Shortliffe, 1977)