I'm not sure that's too diff. from what I'm saying.

The interesting question is what does the brain use as its general class model 
against wh. to compare new individuals? It's unlikely to be a or the first 
individual face/object as you seem to be suggesting.

Another factor here is that you interpret all these objects with your body - 
you understand other faces and bodies by projecting your own body into them - a 
remarkable example of that is the ability of a c. 2 month old infant to imitate 
the mouth movements of parents (& remember it hasn't seen its own mouth yet).


From: deepakjnath 
Sent: Monday, July 26, 2010 8:38 PM
To: agi 
Subject: Re: [agi] How do we hear music


Okay Mike, 

Let me write down my theory of this phenomenon. my intuition is that brain 
learns in steps and deltas. The brain takes in a fixed amount of only new 
information at a time. So when a person who doesn't have too much impressions 
(image memories) of a chinese person sees a chinese, He takes in the round face 
and the eyes etc which are new info to the seer.

When the seer sees another chinese person the older chinese persons image comes 
back into the working memory. The new person is stored as delta of the other 
person.

As the seer sees more and more people the basic structure is no longer new. The 
new features that get captured become the subtle variations from the basic 
structure. This ability to identify new information becomes a crucial function 
of the brain. Thus as time passes with images of chinese people, the seer will 
be able to capture subtle variation and recognize the person.

People who are not musically trained find it difficult to distinguish between 
notes. But repeated listening to the notes engrave the structure of notes to 
the memory. And complex and subtle variations of the notes become apparent to 
the listener as the base notes are already stored in the memory and so no 
longer new.

cheers,
Deepak




On Tue, Jul 27, 2010 at 12:54 AM, Mike Tintner <[email protected]> wrote:

  Deepak,

  No it's basically a distraction from the problem.  With time and closer 
inspection, they will all look different.

  Correction, it IS useful. It probably tells us something about how the brain 
and an AGI must work 
  First you start with a round blob shape for a class of objects - a face blob, 
and then you refine it and refine it, add more and more detail, for different 
individuals.

  What makes Chinese difficult to individuate at first, is they have a 
particular characteristic wh. would be highly distinctive for a Western 
individual - relatively slanted eyes.  Imagine if a new race all had square 
jaws. You can't take your eyes off that feature at first. With time you learn 
to make adjustments for it, and notice the individual characteristics within 
the narrower eyes.  Ditto elephants are hard to individuate at first because 
they all have these massively distinctive features of huge ears and trunks.

  You start general, and gradually individuate - but you have to individuate - 
your life depends on being able to distinguish individual characteristics as 
well as general forms.


  From: deepakjnath 
  Sent: Monday, July 26, 2010 7:56 PM
  To: agi 
  Subject: Re: [agi] How do we hear music


  Mike,

  All chinese look the same for me. But for a chinese person they don't. Why is 
this? Is there another clue here?

  Thanks,
  Deepak


  On Mon, Jul 26, 2010 at 9:10 PM, Mike Tintner <[email protected]> 
wrote:

    David,

    There must be a fair amount of cog sci/AI analysis of all this -  of how 
the brain analyses and remembers tunes  - and presumably leading theories (as 
for vision). Do you or anyone know more here?

    Also, you have noted something of extreme importance, wh. is a lot more 
than "a step further".

    OTOH you've been analysing how we recognize the same, general tune in 
different, individual renditions. 

    OTOH you've pointed out, we also recognize the INDIVIDUAL differences 
of/variatiions on the same genre/class - we appreciate the different ways 
Davis/Gillespie play as well as that they're playing the same tune.

    Now correct me but isn't the individual dimension of images of all kinds, 
almost entirely missing from AI? The capacity to recognize what makes 
individuals of a species individual, and not just that they belong to the same 
species.  Isn't visual object recognition for example almost entirely focussed 
on recognizing general objects rather than individual objects - that that's an 
example of a general doll, rather than an individual particularly beaten up, or 
just slightly and disturbingly altered doll?

    No doubt AI can recognize individual fingerprints, but it's the capacity to 
recognize individuals as variations on the general - to recognize that he has a 
particularly sarcastic smile, or she has a particularly lyrical, fluid walk,  
or that that tune contrasts harmonious and discordant music (as per rap) in a 
distinctive way - that's missing, no?



    From: David Butler 
    Sent: Monday, July 26, 2010 3:44 PM
    To: agi 
    Subject: Re: [agi] How do we hear music


    When we listen to music there are many elements that come into play that 
create our memory of how the song goes.  If you take a piece of instrumental 
music,  you have the melody, a succession of tones in a certain order,  
duration of each note in the melody,  timbre, or tonal quality, (guitar vs 
trombone), time, how fast the song is played.  Phrasing, what part of the 
melody is emphasized using volume, change of tone quality etc...  Is the melody 
played slurred with all the notes run together or staccato played with short 
notes.   

    Too take it a step further how do we recognize a solo played by Miles Davis 
rather than Dizzy Gillespie  playing the same song both on trumpet but sound 
completely different in style.  How do we recognize when two different 
conductors direct the same music with the same orchestra but yet make it sound 
different?

    .


    On Thu, Jul 22, 2010 at 3:05 PM, Matt Mahoney <[email protected]> wrote:

      deepakjnath wrote:


      > Why do we listen to a song sung in different scale and yet identify it 
as the same song.?  Does it have something to do with the fundamental way in 
which we store memory?


      For the same reason that gray looks green on a red background. You have 
more neurons that respond to differences in tones than to absolute frequencies.

       
      -- Matt Mahoney, [email protected] 





--------------------------------------------------------------------------
      From: deepakjnath <[email protected]>
      To: agi <[email protected]>
      Sent: Thu, July 22, 2010 3:59:57 PM
      Subject: [agi] How do we hear music


      Why do we listen to a song sung in different scale and yet identify it as 
the same song.?  Does it have something to do with the fundamental way in which 
we store memory?

      cheers,
      Deepak

            agi | Archives  | Modify Your Subscription  

            agi | Archives  | Modify Your Subscription  



          agi | Archives  | Modify Your Subscription   

          agi | Archives  | Modify Your Subscription  




  -- 
  cheers,
  Deepak

        agi | Archives  | Modify Your Subscription   

        agi | Archives  | Modify Your Subscription  




-- 
cheers,
Deepak

      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/
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=8660244&id_secret=8660244-6e7fb59c
Powered by Listbox: http://www.listbox.com

Reply via email to