ZOMBIE FORUMS

It's a stinking, shambling corpse grotesquely parodying life.
It is currently Thu Mar 28, 2024 2:26 pm

All times are UTC - 8 hours [ DST ]




Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 13 posts ] 
Author Message
 Post subject: Mind Mapped onto Machine Mechanics
PostPosted: Fri Dec 31, 2004 4:54 pm 
Offline
Addict
User avatar

Joined: Fri Jun 27, 2003 6:10 pm
Posts: 2571
Tamayo wrote:
In that brains are neural networks (whatever else they are), and that consciousnesses are hosted by brains, and that neural networks are very difficult to program directly, I suspect even that it will be centuries at least before a person's consciousness can be hosted on a machine that is not that person's original brain.


The Baron wrote:
hey, lookie there! an article on this by a guy who teaches at my goddamn school that apparently won't let me check my grades, fix my registration for next semester, or pay them! </bitter>

good article. seems to emphasize one of the things I was about to mention before I looked this up, that the brain is massively parallel. it's keeping all autonomic functions running, parsing sensory information from an incredible number of sources, and then all higher level functions (aka, sentience). I'm inclined to agree with Tamayo on that it will be centuries before someone's consciousness can be mapped into a machine; we have to understand the brain in its entirety before we can even really start researching that line of thought. I know exactly zero about AI research, so I'll let Tamayo talk about that if she wants.


Thinman wrote:
The rest of this post is more AI stuff, and probably only interesting if you're into that kind of thing.

The thing is, most AI research (that I'm aware of, anyway) doesn't even bother trying to emulate how the human brain works; instead it focuses on ideas that can be rigorously mathematically understood. For one thing, the brain is an ad hoc system. It's a product of millions of years of accretion of functionality. Things are hopelessly interconnected and not always even done any particularly optimal way. For another, biologists and psychoperceptual scientists keep changing their theories of how the mind works.

Our little friend the <a href="http://en.wikipedia.org/wiki/Neural_network">multilayer perceptron</a> was a something of a scientific fad. Claims about great results emulating nature with "neural networks" have given way to the general conclusion that MLPs are delicate, temperamental, algorithms to set up, difficult to analyze once operational and anyway they bear an only superficial resemblance to how (we currently think) the brain actually works.

The current hot topic is kernel methods, like the <a href="http://en.wikipedia.org/wiki/Support_vector_machine">support vector machine</a> which reduces the learning problem to an optimization that can be solved with deterministic numerical methods. It's stable, general, works very well, and bears no resemblance to anything known in nature.[1]




1. Vapnik, the primary mathematician behind the statistical methods used in the SVM, has written a hilariously bitter critique of the neural network community, who ignored his superior theoretical work for many years in favor of the "more natural" MLP.


Emy wrote:
Thinman wrote:
The rest of this post is more AI stuff, and probably only interesting if you're into that kind of thing.

The thing is, most AI research (that I'm aware of, anyway) doesn't even bother trying to emulate how the human brain works; instead it focuses on ideas that can be rigorously mathematically understood. For one thing, the brain is an ad hoc system. It's a product of millions of years of accretion of functionality. Things are hopelessly interconnected and not always even done any particularly optimal way. For another, biologists and psychoperceptual scientists keep changing their theories of how the mind works.

Our little friend the <a href="http://en.wikipedia.org/wiki/Neural_network">multilayer perceptron</a> was a something of a scientific fad. Claims about great results emulating nature with "neural networks" have given way to the general conclusion that MLPs are delicate, temperamental, algorithms to set up, difficult to analyze once operational and anyway they bear an only superficial resemblance to how (we currently think) the brain actually works.

The current hot topic is kernel methods, like the <a href="http://en.wikipedia.org/wiki/Support_vector_machine">support vector machine</a> which reduces the learning problem to an optimization that can be solved with deterministic numerical methods. It's stable, general, works very well, and bears no resemblance to anything known in nature.[1]

1. Vapnik, the primary mathematician behind the statistical methods used in the SVM, has written a hilariously bitter critique of the neural network community, who ignored his superior theoretical work for many years in favor of the "more natural" MLP.
What ever happened to colloid physics/chemistry?

_________________
-DNI ~ by Ezelek
I have earned the title of Pedant.

Image


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 5:13 pm 
Offline
Addict
User avatar

Joined: Fri Nov 08, 2002 5:00 pm
Posts: 4439
Location: You can't take the sky from me. Since I found Serenity.
You should have gotten a mod to split the topic.

Read GEB if you haven't.

Also, curent CPUs don't do actual multi-tasking. It flitters between tasks every so often if it's doing two things at once.

Also, a large part of the brains functions relies on the structure brain itself (patterns of neurons etc). It's hardware based rather than software based. Which means that the CPU would have to emulate the brain. Making it even slower.

As for building a thinking machine from scratch. That strikes me as inordinatly complex. But I'm not going to dismiss the notion off hand. But I will say that in order for it to become a reality by 2020 as the article claims, they will have to go the same route as Deep Blue and design specialized chips for things like sensory input to save the CPU for the hard crunching, like interpreting sentences.

_________________
Build a man a fire, warm him for a day,
Set a man on fire, warm him for the rest of his life.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 5:24 pm 
Offline
Addict
User avatar

Joined: Fri Jun 27, 2003 6:10 pm
Posts: 2571
OmnipotentEntity wrote:
You should have gotten a mod to split the topic.

I am a mod... who doesn't exactly know how to move posts and topics. >__>

But I didn't want to split the thread so that it'd be all messy. This way, ti's everywhere. And the important parts of the posts relivant to this were clearly posted. I don't know. I'm fucking around because I have nothing else to do.

_________________
-DNI ~ by Ezelek
I have earned the title of Pedant.

Image


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 5:48 pm 
Offline
Addict
User avatar

Joined: Mon Aug 09, 2004 2:44 pm
Posts: 1821
Location: Home! Wheeeeee!
This isn't really a debate, i think, it was just off-topic to the other thread.

Oh well. no worries, right?

_________________
Image


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 5:52 pm 
Offline
User avatar

Joined: Sun Jan 17, 2038 9:00 pm
Posts: 3209
Quote:
Also, curent CPUs don't do actual multi-tasking. It flitters between tasks every so often if it's doing two things at once.

CPUs don't, GPUs (which become more general with every generation) do. multi-core chips move CPUs in a more GPU-like direction. silly, eh?


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 7:10 pm 
Offline
Addict
User avatar

Joined: Fri Nov 08, 2002 5:00 pm
Posts: 4439
Location: You can't take the sky from me. Since I found Serenity.
The Baron wrote:
Quote:
Also, curent CPUs don't do actual multi-tasking. It flitters between tasks every so often if it's doing two things at once.

CPUs don't, GPUs (which become more general with every generation) do. multi-core chips move CPUs in a more GPU-like direction. silly, eh?


Almost as silly as my spelling of current.

_________________
Build a man a fire, warm him for a day,
Set a man on fire, warm him for the rest of his life.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 7:33 pm 
Offline
Addict
User avatar

Joined: Thu Feb 06, 2003 4:58 pm
Posts: 3672
Location: Twin Cities, MN
OmnipotentEntity wrote:
The Baron wrote:
Quote:
Also, curent CPUs don't do actual multi-tasking. It flitters between tasks every so often if it's doing two things at once.

CPUs don't, GPUs (which become more general with every generation) do. multi-core chips move CPUs in a more GPU-like direction. silly, eh?


Almost as silly as my spelling of current.


And perhaps as silly as the thread title?

Mind Mapped Onto Machine Mechanics. Say that five times fast.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Dec 31, 2004 10:39 pm 
Offline
Addict
User avatar

Joined: Fri Jun 27, 2003 6:10 pm
Posts: 2571
Wark wrote:
OmnipotentEntity wrote:
The Baron wrote:
Quote:
Also, curent CPUs don't do actual multi-tasking. It flitters between tasks every so often if it's doing two things at once.

CPUs don't, GPUs (which become more general with every generation) do. multi-core chips move CPUs in a more GPU-like direction. silly, eh?


Almost as silly as my spelling of current.


And perhaps as silly as the thread title?

Mind Mapped Onto Machine Mechanics. Say that five times fast.


>.> I think alliteration is fun.
Teehee.

_________________
-DNI ~ by Ezelek
I have earned the title of Pedant.

Image


Top
 Profile  
 
 Post subject: I like this topic. Can you tell?
PostPosted: Sat Jan 01, 2005 12:01 am 
Offline
Addict
User avatar

Joined: Thu Sep 04, 2003 9:43 pm
Posts: 1096
Emy wrote:
What ever happened to colloid physics/chemistry?

Huh? What does colloidal matter have to do with AI?
Something to do with stochastic methods? (Simulated annealing?)

(I'm not a real AI person. I've just dabbled in pattern recognition a bit)

OmnipotentEntity wrote:
But I will say that in order for it to become a reality by 2020 as the article claims, they will have to go the same route as Deep Blue and design specialized chips for things like sensory input to save the CPU for the hard crunching, like interpreting sentences.

Too right. Human-style visual perception is incredibly complex. It takes a large portion of your brain to extract textures, find edges and shapes, convert the 2D information from your eyes into a perception of 3D, and recognize objects.

You could build the vision system of a simpler animal like a frog with much less effort, but what you'd have in the end would be not much more than a prey / notprey detector. (Predators usually have both better vision and more intelligence than their prey and there is a school of thought that associates the evolution of a more sophisticated visual perception system with the emergence of higher intelligence.)

_________________
Always watching, ever vigilant


Top
 Profile  
 
 Post subject:
PostPosted: Sat Jan 01, 2005 4:11 am 
Offline
Addict
User avatar

Joined: Wed Aug 13, 2003 2:39 am
Posts: 1756
Location: The border of civilization
Who says that the robot (that's what you peps are talking about here) needs a 3D vision with colour? Give the thing a sonar, even today a PC can make a 3D map by sonar readings with the right software.

_________________
Warning! The owner of this property is armed and willing to defend life, liberty and property.


Top
 Profile  
 
 Post subject:
PostPosted: Sat Jan 01, 2005 11:39 am 
Offline
Addict
User avatar

Joined: Fri Nov 08, 2002 5:00 pm
Posts: 4439
Location: You can't take the sky from me. Since I found Serenity.
Nothing say that this *has* to happen. However, sonar is only viable for night time situations. Much more can be disterned from light than sonar. Plus sonar has an annoying habit of going around corners and through objects, screwing you up.

And more things emit sound than light. Again fucking with your perception.

I wouldn't argue against both. But visual input is a definate must.

_________________
Build a man a fire, warm him for a day,
Set a man on fire, warm him for the rest of his life.


Top
 Profile  
 
 Post subject:
PostPosted: Sat Jan 01, 2005 11:57 am 
Offline
Addict
User avatar

Joined: Thu Sep 04, 2003 9:43 pm
Posts: 1096
Gazing Rabbit wrote:
Who says that the robot (that's what you peps are talking about here) needs a 3D vision with colour? Give the thing a sonar, even today a PC can make a 3D map by sonar readings with the right software.

This is true. You can eliminate the 2D to 3D mapping problem by using any number of different imaging systems. (SONAR, ranging laser, etc) But any vision system more advanced than a collision avoider still has the problem of mapping that information from a static (N-dimensional) image into a perceived environment. This leaves you with the big problems of object recognition (Is object A a bottle or a can?) and scene analysis (Is there a can in the current field of view?)

Really, we're talking about the difference between vision and visual perception. You see with the back portion of your brain, not your eyes.

_________________
Always watching, ever vigilant


Top
 Profile  
 
 Post subject:
PostPosted: Mon Jan 03, 2005 1:35 pm 
Offline
Addict
User avatar

Joined: Fri Nov 08, 2002 5:00 pm
Posts: 4439
Location: You can't take the sky from me. Since I found Serenity.
Well, that and those 3D sonar mapping things are utilized primarily for things that don't change much.

Like rocks.

In a preditor/prey aspect not only are you giving your posistion away every time you ping for location, but you are wasting a split second.

_________________
Build a man a fire, warm him for a day,
Set a man on fire, warm him for the rest of his life.


Top
 Profile  
 
Display posts from previous:  Sort by  
Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 13 posts ] 

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 16 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group