XPost: comp.ai.fuzzy, sci.logic, sci.cognitive   
   From: Dr.PeiWang@gmail.com   
      
   David Kinny wrote:   
   > "Glen M. Sizemore"  writes:   
   >   
   > >"PeiWang"  wrote in message   
   > >news:45317299$1@news.unimelb.edu.au...   
   > >> New Book Announcement [apologies for cross-posting]   
   > >>   
   > >> Rigid Flexibility: The Logic of Intelligence   
   > >> by Pei Wang   
   > >> Springer, October 2006, ISBN: 1402050445   
   > >>   
   > >> This book provides the blueprint of a thinking machine.   
   > >>   
   > >> While most of the current works in Artificial Intelligence   
   > >> (AI) focus on individual aspects of intelligence and   
   > >> cognition, the project described in this book, Non-Axiomatic   
   > >> Reasoning System (NARS), is designed and developed to attack   
   > >> the AI problem as a whole.   
   > >>   
   > >> This project is based on the belief that what we call   
   > >> "intelligence" can be understood and reproduced as "the   
   > >> capability of a system to adapt to its environment while   
   > >> working with insufficient knowledge and resources". According   
   > >> to this idea, a novel reasoning system is designed, which   
   > >> challenges all the dominating theories in how such a system   
   > >> should be built. The system carries out reasoning, learning,   
   > >> categorizing, planning, decision making, etc., as different   
   > >> facets of the same underlying process.   
   >   
   > >Sorry, but this unification has already been suggested by a philosophy known   
   > >as radical behaviorism. See Skinner (1938, 1945, 1953, 1957, 1960, etc. etc.   
   > >etc. etc. etc.).   
   >   
   > So what?  Perhaps Wang takes Skinner as one of his inspirations :)   
   >   
   > When described at a high-level, quite different approaches can sound   
   > very similar. But a quick glance at the TOC shows that much of the book   
   > content is some kind of logic and an associated reasoning system,   
   > i.e. fairly GOFAI, so whatever the value of Wang's contribution, its   
   > intersection with Skinner's contribution is likely to be very small.   
   >   
   > David   
   >   
      
   The unification of mind/cognition/intelligence has been proposed by   
   many people, though each of them has suggested a very different   
   approach to reach it.   
      
   I got much less inspiration from Skinner than from Piaget.   
      
   At the first glance, my approach does look like GOFAI, but if you look   
   into it, you'll see that I've challenged most of its key notions, like   
   "representation", "validity", "consistency", "algorithm", "complexity",   
   and so on. The philosophical foundation of my work is actually much   
   closer to connectionism than to GOFAI, though the "neural network"   
   technique is too limited for my goal. Many people won't believe that   
   when designed properly, a reasoning system can be more flexible than a   
   NN in many aspects.   
      
   Pei   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)