Ben, You are making a v. strong claim here. If you can back it up with:
a) a general text exposition so that even a techno-idiot like me (& non-technical psychologists/ philosophers) can understand b) a technical exposition c) that magical video demo you have your breakthrough - & sufficient to go public, a la Hawkins - and claim the first real example of AGI. Even if it has only limited generality and robustness, it would be something. I'm inclined to be sceptical & suspect that you have "led the witness." There isn't enough info in your post - enough of an "idea" - (esp. of how it draws analogies) - for me. But you certainly have me paying attention. And I (& I hope others) would welcome still more info. Ben wrote: MT:Sounds like a good analogy. If it can play fetch, it can play hide-and-seek. [And exactly the sort of thing that a true AGI must do - absolute heart of AGI].The question, wh. I wouldn't think that complex to answer, is: how did it connect the action/activity of fetch, to the activity of hide-and-seek? (And that connection surely has nothing to do, essentially, with probability). [My first impression is that that's a graphic/imaginative analogy. It certainly would be, I'd argue, if a human or child made it]. Did the system/pet draw the analogy itself, without being told? -- The way this connection is made in Novamente actually has a lot to do with probability ;-) -- There is actually an imagistic aspect, and a non-imagistic aspect -- The non-imagistic aspect: a bunch of probabilistically-weighted semantic nodes and links are created related to hide-and-seek (based what the system observes about the game via watching others play it, and via getting reinforcement from others about how well it plays it). These sets of nodes and links are then compared using probabilistic logic, including inductive and abductive inference. -- The imagistic aspect: fetch and hide-and-seek may be enacted in the system's "internal simulation world" (which at the moment is a pretty simple 2D simulation-world beast, but will become more fully featured later on), each yielding a set of nodes and links called a "behavior description" which may then be manipulated using probabilistic logic (So in a sense there is some visual imagination here -- there's a "mind's eye" simulation of the world, yet this is directly tied into the declarative and procedural knowledge stores which are manipulated probabilistically.) -- yes, the system draws analogies like this itself, without being told -- according to what I have read, not all humans draw analogies like this visually and imagistically. Some do, some don't. (speaking of reading material, you should look at Hadamard's book The Psychology of Mathematical Invention. He interviewed a load of brilliant mathematicians about their cognitive styles and heuristics. There is a lot of diversity. A lot of visual/sensory thinkers, and a lot who aren't.) -- Ben G ------------------------------------------------------------------------------ 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/?&; ------------------------------------------------------------------------------ No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.488 / Virus Database: 269.14.8/1064 - Release Date: 11/10/2007 15:09 ----- 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=53173778-944ec4
