Received: from mit-oz by mit-htvax with CHAOS; 23 Apr 1984 12:41:34-EST Received: from MIT-MC by MIT-OZ via Chaosnet; 23 Apr 84 12:40-EST Date: 23 Apr 1984 12:32 EST (Mon) Message-ID: From: "Daniel S. Weld" To: *bboard%MIT-OZ@MIT-MC.ARPA Subject: AI Revolving Seminar Wednesday 4/25 4:00pm 8th floor playroom The Copycat Project: An Experiment in Nondeterminism and Creative Analogies Doug Hofstadter A micro-world is described, in which many analogies involving strikingly different concepts and levels of subtlety can be made. The question "What differentiates the good ones from the bad ones?" is discussed, and then the problem of how to implement a computational model of the human ability to come up with such analogies (and to have a sense for their quality) is considered. A key part of the proposed system, now under development, is its dependence on statistically emergent properties of stochastically interacting "codelets" (small pieces of ready-to-run code created by the system, and selected at random to run with probability proportional to heuristically assigned "urgencies"). Another key element is a network of linked concepts of varying levels of "semanticity", in which activation spreads and indirectly controls the urgencies of new codelets. There is pressure in the system toward maximizing the degree of "semanticity" or "intensionality" of descriptions of structures, but many such pressures, often conflicting, must interact with one another, and compromises must be made. The shifting of (1) perceived boundaries inside structures, (2) descriptive concepts chosen to apply to structures, and (3) features perceived as "salient" or not, is called "slippage". What can slip, and how, are emergent consequences of the interaction of (1) the temporary ("cytoplasmic") structures involved in the analogy with (2) the permanent ("Platonic") concepts and links in the conceptual proximity network, or "slippability network". The architecture of this system is postulated as a general architecture suitable for dealing not only with fluid analogies, but also with other types of abstract perception and categorization tasks, such as musical perception, scientific theorizing, Bongard problems and others. ************************************************************************** 12:00 NOON 8TH FLOOR PLAYROOM FRIDAY 5/5 GRADUAL STUDENT LUNCH SEMINAR SERIES The G0001 Project: An Experiment in G0002 and Creative G0003 A G0004 is described, in which many G0003 involving strikingly different G0005 and levels of G0006 can be made. The question "What differentiates the good G0003 from the bad G0003?" is discussed, and the problem of how to G0008 a G0009 G0010 of the G0011 G0012 to come up with such G0003 (and to have a sense for their quality) is considered. A key part of the proposed system, now under development, is its dependence on G0013 G0014 G0015 of G0016 interacting "G0017" (selected at random to G0019 with G0020 proportional to G0021 assigned "G0022"). Another key G0023 is a G0024 of linked G0005 of varying levels of "G0025", in which G0026 spreads and G0027 controls the G0028 of new G0017. The shifting of (1) G0033 G0034 inside structures, (2) descriptive G0005 chosen to apply to G0030, and (3) G0043 perceived as "G0031" or not, is called "G0032". What can G0031, and how, are G0014 G0033 of the interaction of (1) the temporary ("G0034") structures involved in the G0003 with (2) the permanent ("G0035") G0005 and links in the G0036 network, or "G0037 network". The G0038 of this system is G0039 as a general G0038 suitable for dealing not only with fluid G0003, but also with other types of G0039 G0040 and G0041 tasks, such as musical G0040, G0041 G0042, Bongard problems and others. Hosts: Harry Voorhees and Dave Siegel ****************************************************************