Probabilistic Logic Networks A Comprehensive Framework for Uncertain
This book describes Probabilistic Logic Networks (PLN), a novel conceptual, mathematical and computational approach to uncertain inference. Going beyond prior probabilistic approaches to uncertain inference, PLN encompasses such ideas as induction, abduct
- PDF / 31,740,497 Bytes
- 330 Pages / 439.37 x 666.142 pts Page_size
- 8 Downloads / 196 Views
Probabilistic Logic Networks A Comprehensive Framework for Uncertain Inference
Ben Goertzel Matthew Iklé Izabela Freire Goertzel Ari Heljakka
13
Authors Ben Goertzel Novamente LLC 1405 Bernerd Place Rockville, MD 20851 [email protected]
Matthew Iklé Department of Chemistry, Computer Science, and Mathematics Adams State College Alamosa, CO 81102 [email protected]
Izabela Freire Goertzel Novamente LLC 1405 Bernerd Place Rockville, MD 20851 [email protected]
ISBN: 978-0-387-76871-7 DOI: 10.1007/978-0-387-76872-4
Ari Heljakka Novamente LLC 1405 Bernerd Place Rockville, MD 20851 [email protected]
e-ISBN: 978-0-387-76872-4
Library of Congress Control Number: 2008924650 © 2008 Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper springer.com
Dedicated to the memory of Jeff Pressing
Physicist, psychologist, musician, composer, athlete, polyglot and so much more -Jeff was one of the most brilliant, fascinating, multidimensional and fully alive human beings any of us will ever know. He was killed in his sleep by a sudden, freak meningitis infection in 2002, while still young and in perfect health, and while in the early stages of co-developing the approach to probabilistic reasoning described in this book. Jeff saw nearly none of the words of this book and perhaps 25% of the equations. We considered including him as a posthumous coauthor, but decided against this because many of the approaches and ideas we introduced after his death are somewhat radical and we can’t be sure he would have approved them. Instead we included him as co-author on the two chapters to whose material he directly contributed. But nonetheless, there are many ways in which the overall PLN theory presented here – with its combination of innovation, formality and practicality -- embodies Jeff’s “spirit” as an intellect and as a human being. Jeff, we miss you in so many ways!
Contents 1.
Introduction
1
2.
Knowledge Representation
23
3.
Experiential Semantics
41
4.
Indefinite Truth Values
49
5.
First-Order Extensional Inference: Rules and Strength Formulas (Coauthored with Jeff Pressing) 63
6.
First-Order Extensional Inference with Indefinite Truth Values
7.
First-Order Extensional Inference with Distributional Truth Values 141
8.
Error Magnification in Inference Formulas
149
9.
Large-Scale Inf
