======== WDL Dis. #8 (Bcc: all WDL participants) ====== >From franklin@anduril.memphis.edu Sun Feb 27 17:20:21 2000 To: Dr John J Weng Subject: Re: WDL Dis. #6 John, > Are you aware of any computational model that realizes > this kind of "moving down" process through automatic development? Chunking in SOAR is such a computational mechanism, the first of which I'm aware. [Laird, John E., Newell, Allen, and Rosenbloom, Paul S. (1987). "SOAR: An Architecture for General Intelligence." Artificial Intelligence, 33: 1-64.] The formation of concept demons in Jackson's pandemonium theory is a second such. [Jackson, John V. (1987), "Idea for a Mind," SIGGART Newsletter, no. 181, July, 23-26.] I've described both of these in my *Artificial Minds* [MIT Press, 1995]. We've implemented a version, called concept codelets, of the second in our "conscious" software agents. It's described in the paper I've submitted to this conference [Learning in "Conscious" Software Agents] in the section called Temporal Proximity Learning. In the two agents discussed there the construction hasn't yet proved of use. I suspect this is a result of the simple domain of the first, and to the second not yet running so we can't see as yet what it will do. I believe this kind of learning will prove to of critical importance when such software agents routinely go through development cycles as suggested in the abstract to my viewpoint talk. Stan -- Stan Franklin Math Sciences Dept Phone: (901) 678-3142 Univ of Memphis Fax: (901) 678-2480 Memphis, TN 38152 stan.franklin@memphis.edu USA www.msci.memphis.edu/~franklin