The Limits of Artificial Intelligence (Classic Reprint) - Softcover

Schwartz, Jacob T.

9781332151523: The Limits of Artificial Intelligence (Classic Reprint)

Softcover

ISBN 10: 1332151523 ISBN 13: 9781332151523

Verlag: Forgotten Books, 2018

Alle Exemplare dieser ISBN-Ausgabe

0 Gebraucht

2 Neu

Von EUR 18,85

Excerpt from The Limits of Artificial Intelligence

This philosophical debate concerns the existence of fundamental limits to the artificial intelligence enterprise, which however is only one of several kinds of potentially significant limit that need to be considered. Even if no such fundamental limits existed, i.e. Even if a hypothetical infinitely fast computing engine possessed of infinite amounts of memory could in principle duplicate all aspects of human mental capability, it would still remain necessary to ask just how much computation and data storage such duplication would require. Suppose, for example, that it could be shown that the minimum computational resource required to duplicate some human mental function is implausibly large, relative either to the extreme limits of physically realizable computation, or to the largest computers likely to be constructed over the next decades or centuries. In this case, construction of significant artificial intelligences would be blocked by inescapable practicallimits, even if fundamental limits did not exist. Finally, even if no such computational factors proved to limit the possibility of artificial intelligence, one would still want to assess the existing state of the field and project the rate of progress likely to result from application of its present intellectual tools to the profound problems with which it must wrestle.

The next five sections of the present article develop points relevant to the three kinds of limits defined in the preceding paragraph. A final section discusses certain other concerns, implicit in the debate between the enthusiasts of artificial intelligence and their opponents, which may explain some of the vehemence which has crept into this debate.

About the Publisher

Forgotten Books publishes hundreds of thousands of rare and classic books. Find more at www.forgottenbooks.com

This book is a reproduction of an important historical work. Forgotten Books uses state-of-the-art technology to digitally reconstruct the work, preserving the original format whilst repairing imperfections present in the aged copy. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in our edition. We do, however, repair the vast majority of imperfections successfully; any imperfections that remain are intentionally left to preserve the state of such historical works.

Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.

Rese�a del editor

Excerpt from The Limits of Artificial Intelligence

The question of what intrinsic limits constrain the artificial intelligence enterprise, which can be defined as the attempt to construct electronic systems exhibiting human or superhuman levels of capability in areas traditionally regarded as mental, has been debated within very wide limits. On one side one finds a substantial community of researchers who believe firmly that such systems will prove possible. Their common (but not universal) assumption is that the organic brain is in effect a complex electrochemical system operating in some (doubtless highly parallel) but essentially computer-like fashion, and hence gives direct proof of the realizability of intelligence by mechanism; vide Marvin Minsky's flat-footed 'The brain is a meat machine'. Opposing this view one finds the assertion that mental processes are essentially indecomposable, lie outside the narrow reach of scientific reductionism, and that their indecomposability sets fundamental limits to any attempt to duplicate intelligence by mechanism. From this point of view, e.g. as represented by the writings of Hubert Dreyfus, the history of artificial intelligence research to date, consisting always of very limited success in particular areas, followed immediately by failure to reach the broader goals at which these initial successes seem at first to hint, gives empirical proof of the presence of irreducible wholes fundamentally incapable of being comprehended, much less duplicated, by the narrowly technical procedures of artificial intelligence researchers.

This philosophical debate concerns the existence of fundamental limits to the artificial intelligence enterprise, which however is only one of several kinds of potentially significant limit that need to be considered. Even if no such fundamental limits existed, i.e. even if a hypothetical infinitely fast computing engine possessed of infinite amounts of memory could in principle duplicate all aspects of human mental capability, it would still remain necessary to ask just how much computation and data storage such duplication would require. Suppose, for example, that it could be shown that the minimum computational resource required to duplicate some human mental function is implausibly large, relative either to the extreme limits of physically realizable computation, or to the largest computers likely to be constructed over the next decades or centuries.

About the Publisher

Forgotten Books publishes hundreds of thousands of rare and classic books. Find more at www.forgottenbooks.com

This book is a reproduction of an important historical work. Forgotten Books uses state-of-the-art technology to digitally reconstruct the work, preserving the original format whilst repairing imperfections present in the aged copy. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in our edition. We do, however, repair the vast majority of imperfections successfully; any imperfections that remain are intentionally left to preserve the state of such historical works.

��ber diesen Titel� kann sich auf eine andere Ausgabe dieses Titels beziehen.