Response to Mike Tintners Thu 10/4/2007 7:36 AM post: I skimmed LGIST: Learning Generalized Image Schemas for Transfer Thrust D Architecture Report, by Carole Beal and Paul Cohen at the USC Information Sciences Institute. It was one of the PDFs listed on the web link you sent me (at http://eksl.isi.edu/files/papers/cohen_2006_1160084799.pdf). It was interesting and valuable. I found its initial few pages a good statement of some solid AI ideas. Its idea of splitting states based on entropy is a good one, one that I have myself have considered as a guide for when and where in semantic space to split models and how to segment temporal representations.
But the system appears to be a pretty small one. It appears to start out with a fair number of relatively abstract programmer-defined concepts (shown in Table One), which is probably necessary for what it is trying to accomplish on the hardware it is trying to accomplish it on. But is very different than a more human brain approach, that would be able to learn most of those concepts itself, and thus would probably have them be more grounded. It starts with a lot of the hard work necessary for the type of problem it is trying to solve already done (which is very helpful for what it is trying to do), but which at least suggests a question about how well it will be able to learn in areas where that type of hard work has not already been done for it. Edward W. Porter Porter & Associates 24 String Bridge S12 Exeter, NH 03833 (617) 494-1722 Fax (617) 494-1822 [EMAIL PROTECTED] -----Original Message----- From: Mike Tintner [mailto:[EMAIL PROTECTED] Sent: Thursday, October 04, 2007 7:36 AM To: agi@v2.listbox.com Subject: [agi] Another AGI Project Another AGI project -some similarities to Ben's. (I was not however able to play with my Wubble - perhaps you'll have better luck). Comments? http://eksl.isi.edu/cgi-bin/page.cgi?page=project-jean.html ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=8660244&id_secret=49688212-d1bb83
