I would venture to say that the child brings to bear a known hand shape, not a new hand shape. Then iteratively adjusts this shape as needed. A la Cased based reasoning: retrieve a known solution to the present situation, then adapt the solution to the specific situation. ~PM.
From: [email protected] To: [email protected] Subject: Re: [agi] Does Siri + Watson + Cleverbot = AGI ? Date: Mon, 1 Oct 2012 01:56:50 +0100 Agreed. I didn’t say it adjusts “straight away” – groping can involve many successive attempts. I said it forms an initial hand shape straight away - “has a stab at it”. Even that isn’t sacrosanct, of course. But the point is that reality – “having a stab at it” – “adventure” in the technical sense of “venturing ad/towards the goal” – is in total contrast to Jim’s/many AGI-ers’ idea of AGI as involving the careful consideration of massively complex alternatives in order to act. Complexity only applies to routine actions, where the agent already knows how to perform an action, including all the possible alternatives. Creativity/AGI is about attempting NEW actions. From: Piaget Modeler Sent: Monday, October 01, 2012 1:40 AM To: AGI Subject: RE: [agi] Does Siri + Watson + Cleverbot = AGI ? Mike Tinter said: "It forms a tentative new hand shape straight away and adjusts that to fit – it “gropes”" Not exactly Mike. "THE USE OF REFLEXES -- Concerning its adaptation, it is interesting to note that the reflex, no matter how well endowed with hereditary physiological mechanism, and no matter how stable its automatization, nevertheless needs to be used in order truly to adapt itself , and that it is capable of gradual accommodation to external reality." "...but as we know (Obs. 1), it sometimes happens that the child does not adapt at the first attempt. Only practice will lead to normal functioning. That is the first aspect of accommodation: contact with the object modifies, in a way, the activity of the reflex, and , even if this activitiy were oriented hereditarily to such contact, the latter is no less necessary to the consolidation of the former. This is how certain instincts are lost or certain reflexes cease to function normally, due to the lack of a suitable environment. Moreover, contact with the enviornment not only results in developing the reflexes, but also in coordinating them in some way." ~ Jean Piaget, The Origins of Intelligence in Children, pp. 29 -30. I would submit that the child or robot puts forth it's best hand-shape (that closest match to the object), and then successively attempts modified hand shapes according to the degree of success or difficulty encountered with the first attempt. Hence, practice (reinforcement and correction, or as I prefer to say , regulation and compensation) plays a significant part even in the smallest of reflexes. AGI Requirement #243: Reflexes are not fixed, but must be capable of adaptation to new situations. ~PM. From: [email protected] To: [email protected] Subject: Re: [agi] Does Siri + Watson = AGI ? Date: Mon, 1 Oct 2012 00:04:51 +0100 No Jim, it’s totally different. When a child or robot has to handle a new object, it does not have a set of hand shapes which it systematically considers and combines to form the new shape. It forms a tentative new hand shape straight away and adjusts that to fit – it “gropes”. It does not run through an amazing combinatorial explosion of hand shapes. Puh-lease. When a child or robot has to explore a new territory – let’s say a rubbish dump or adventure playground/ obstacle course - it does not run through a combinatorial explosion of possible routes – it doesn’t know the territory! It has to discover the territory and obstacles as it goes along. By the same token, it does not run through a combinatorial explosion of possible obstacles, or possible ways to climb over them. These are new and different obstacles – even if also somewhat similar to previously encountered ones. The whole idea of combinatorial explosion applied to any creative activity/ new journey in a new field/ new composition is *absurd*. What was the combinatorial explosion you went through to compose your one-line post in response to mine?. That was a micro-new-journey. You were confronted with a newish idea, you came up with a newish response if only in context - in no way was that the result of a combinatorial explosion. If you did not have the handicap of being cognitively unable to consider examples of problemsolving , you would realise this more or less immediately. There can be no combinatorial explosion when you pursue a journey one step at a time - here are some real life, real world problemsolving examples: https://www.google.co.uk/search?q=%22have+to+take+it+one+step+at+a+time%22&oq=%22have+to+take+it+one+step+at+a+time%22&sugexp=chrome,mod=4&sourceid=chrome&ie=UTF-8 From: Jim Bromer Sent: Sunday, September 30, 2012 10:59 PM To: AGI Subject: Re: [agi] Does Siri + Watson = AGI ? On Sun, Sep 30, 2012 at 10:12 AM, Mike Tintner <[email protected]> wrote: If you can improvise - you can freely combine and innovate movements of different limbs. That is why we sometimes talk about things like "the combinatorial explosion." "freely combine..." combinatorial... AGI | Archives | Modify Your Subscription AGI | Archives | Modify Your Subscription AGI | Archives | Modify Your Subscription AGI | Archives | Modify Your Subscription ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
