Chen Robert H. - Artificial Intelligence : An Introduction for the Inquisitive Reader

Robert H. Chen

Chelsea Chen

First edition published 2022

by CRC Press

6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press

4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

2022 Robert H. Chen and Chelsea Chen

CRC Press is an imprint of Taylor & Francis Group, LLC

Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under US Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access

Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloguing-in-Publication Data

Names: Chen, Robert H., 1947- author. | Chen, Chelsea, author.

Title: Artificial intelligence : an introduction for the inquisitive reader / authored by Robert H. Chen, Chelsea Chen.

Description: Boca Raton : CRC Press, 2022. | Includes bibliographical references and index.

Identifiers: LCCN 2021055530 (print) | LCCN 2021055531 (ebook) | ISBN 9781032103471 (hardback) | ISBN 9781032101842 (paperback) | ISBN 9781003214892 (ebook)

Subjects: LCSH: Artificial intelligence.

Classification: LCC Q335 .C4845 2022 (print) | LCC Q335 (ebook) | DDC 006.3--dc23/eng/20211116

LC record available at https://lccn.loc.gov/2021055530

LC ebook record available at https://lccn.loc.gov/2021055531

ISBN: 978-1-032-10347-1 (hbk)

ISBN: 978-1-032-10184-2 (pbk)

ISBN: 978-1-003-21489-2 (ebk)

DOI: 10.1201/9781003214892

Typeset in Minion

by MPS Limited, Dehradun

I N THIS BOOK, THE ADVENTURES IN THE QUEST FOR artificial intelligence (AI) are exemplified by the entertaining demonstrations of man versus machine competitions such as IBM Deep Blue versus Garry Kasparov and Google AlphaGo versus Lee Sedol, but the real significance of AI evolves from the human ideas behind the machines algorithms and what the machine may be capable of in the future. Starting with mechanical calculation, the development of artificial intelligence can be seen as a natural progression of technology abetted by the generation of computer science. With the hardware, software, and communications in hand, the quest for the long-dreamed of expert system began in earnest with the pure logic-based top-down machine where axioms go in and mathematical theorems come out, but because of the inherent contradictions in pure logic and lack of data and computer power, the Logic Theorist was replaced by Big Data and massively parallel processing machines, in the so-called bottom-up approach.

Bottom-up AI mimics the structure and pattern recognition capability of human brains with electrically activated artificial neurons forming synaptic patterns of recognition and thought in an artificial neural network formed in a parallel-processing computer. The synaptic patterns are produced by Markov chain modeling of neuron activation with the neurons weighted in accord with a training set in a process called parameterization. The gradient descent of vector calculus minimizes the difference between the machine patterns and the ground truth, backpropagating the differences using the chain rule of calculus provides the machine learning, and hyperparameterization fine-tunes the accuracy and computational efficiency of the machines thinking process.

The example of convolutional neural network used in computer vision armed with Big Data and massively parallel processing has brought machine vision into almost every facet of industry and society.

Predictive analytics AI is employed almost everywhere, especially critically in business, science, politics, and the military. AI reinforcement learning has produced machines that can best the top human video gamers without even a priori knowing the rules of the games being played, and in the real world of imperfect knowledge, AI systems have beaten the best in Texas Holdem poker competition.

Because of the ambiguities and vagaries of speech, natural language processing using generative recurrent neural networks that can assess immediate speech in terms of what was spoken or inferred before are presently the most promising approaches.

The basic ideas behind the implementing algorithms and the processes enabling a machine to learn are described within the theme of the difference between human and artificial intelligence, with the touchstone question being the capability of machines to do mathematics.

That ability required by the reader, however, is only basic calculus, and explanations of the particular equations for artificial intelligence algorithms are provided with examples so that for those who have never learned, or have long forgotten their calculus, the ideas behind the algorithms can be easily understood, hopefully kindling an appreciation of the role of mathematics in artificial intelligence.

As for form, some methods of mathematical exposition, for example the boldface representations of vectors, and curly letter matrices are employed only when necessary for clarity, as their identities are evident from the context. Wording and spelling are in the American style but with British punctuation leaving quotes and parentheses inside sentences of which they are only a part, at the dreadful consequence of laying bare the period or comma.

Only the designs of the algorithms are presented, for those wishing to know the code, referrals are made to the relevant articles, and free code hosting platforms and excellent online programming tutorials offer great convenience and hands-on coding experience.

The authors would like to thank Callum Fraser of Taylor & Francis Publishing for professional guidance, to Mansi Kabra for expert handling of the manuscript, and the reviewers of the manuscript who provided excellent critical suggestions.

DOI: 10.1201/9781003214892-2

H e was a man with two common first names, Arthur Samuel, born and raised in mid-America Emporia Kansas in a middle-class family, a graduate of the local College of Emporia, he appeared to be an ordinary young man brought up in a traditional American society. However, the genial but inherently cautious Arthur Samuel was soon to known as a man of great distinction in an unusual new realm. For from Emporia, his special talents allowed him to enter the citadel of engineering education MIT which provided him a sound electrical engineering basis, and thence to the fount of technical innovation AT&T Bell Labs where he worked on the telecommunications systems that transformed the whole world, and the radar technology critical to victory in World War II.