Re: Temporal Learning
- From: lesterDELzick@xxxxxxxxxxxxxxxx (Lester Zick)
- Date: Tue, 11 Oct 2005 01:26:38 GMT
On 10 Oct 2005 15:03:09 -0700, "feedbackdroids"
<feedbackdroids@xxxxxxxxx> in comp.ai.philosophy wrote:
>
>Lester Zick wrote:
>
>. But you don't show how the steps are
>> necessarily achieved such that they don't result from the identical
>> regression that becomes circular. That's what I mean when I
>> characterize this kind of reduction as homuncular.
>
>
>I think you're getting a little too caught up in equating iterative or
>recursive with homuncular.
I'm getting a little caught up in equating undemonstrated empirical
assumptions of truth and circular arguments with homuncular.
> Homuncular usually has a more specific
>connotation.
The only more specific connotation is the original meaning of little
men in the womb.
> Not at all the same thing or kind, although both types of
>processes happen to involve iteration, but that's where it ends.
Assumptions of truth and circular arguments are always recursive and
iterative but only when people try to argue or follow them.
>When a CS-type talks about writing an iterative or recursive routine,
>they're talking about straight coding, and behaviorist or philosophical
>arguments regards little phantom men have nothing whatsoever to do with
>it.
Unfortunately the lack of proof for assumptions of truth have
everything to do with it or we wouldn't be arguing and talking the
same subjects to death over and over again.
> When a bunch of liquid molecules organize into a Benard cell, it
>has nothing to do with little observers. The molecules go round and
>round, and couldn't care less whether anyone is watching.
And the arguments go round and round as well because no one can
prove anything anyone assumes including the little steps in the step
by step approach to CS. You say ai can only be engineered step by step
but you don't say what the steps have to be such that your steps are
true and others are not.
>You can't
>> demonstrate what the steps have to be that makes them essential to
>> the step by step process you assume versus what others assume.
>>
>> Every historical accumulation of wisdom is step by step. But nobody
>> seems to be able to demonstrate exactly what the steps have to be.
>>
>> ~v~~
>
~v~~
.
- References:
- Re: Temporal Learning
- From: humiguel
- Re: Temporal Learning
- From: Traveler
- Re: Temporal Learning
- From: feedbackdroids
- Re: Temporal Learning
- From: Traveler
- Re: Temporal Learning
- From: Lester Zick
- Re: Temporal Learning
- From: feedbackdroids
- Re: Temporal Learning
- From: Lester Zick
- Re: Temporal Learning
- From: feedbackdroids
- Re: Temporal Learning
- Prev by Date: Re: Temporal Learning
- Next by Date: Re: reward prediction
- Previous by thread: Re: Temporal Learning
- Next by thread: Re: Temporal Learning
- Index(es):
Relevant Pages
|
Loading