U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Science and Technology
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
Reply Start New Thread
 
Old 05-14-2017, 10:28 PM
 
34,409 posts, read 34,520,079 times
Reputation: 13474

Advertisements

Quote:
Originally Posted by davidt1 View Post
-- beating us at chess, jeopardy, and analyzing data thousands of times faster and more accurate than humans ever could.
The only thing that is true AI advancement is beating us at Jeopardy. The first computer could have beat a human at chess, the issue is that it would take forever to calculate the move. Deep Blue beating Kasparov was just a demonstration of raw processing power, at the end of the day chess is just a game of math which is something computers excel at. You can't beat an opponent at chess that can calculate every possible move that are never wrong.

Jeopardy on the other hand is a completely different matter. What is intelligence? What is this? "A"

Seems simple enough but how many different fonts do you see each day some of which you have never seen before that you can immediately recognize as the letter "A". That is what separates us from a regular computer and simple things like this we are doing constantly all the time.

You can feed every conceivable font to computer and tell it that it's the letter A but if try and do that with every piece of human knowledge the task is insurmountable even if you had the ridiculous amount of raw computer power required to do it.

That is where Watson from Jeopardy comes in.You give it 30 different A's and program it to identify other A's by itself. When they ask it question it doesn't have that specific answer stored somewhere. What is important to note is Watson can be wrong. Also important to note is when it is wrong it will adjust it's programming so it's right. Scary huh?

Watson is now Dr. Watson.... they are using it in the medical field to identify abnormalities in cat scans and things of that nature way beyond what a human would be able to identify.

Last edited by thecoalman; 05-14-2017 at 10:40 PM..
Reply With Quote Quick reply to this message

 
Old 05-22-2017, 05:36 PM
 
Location: West Coast of Europe
19,017 posts, read 16,778,707 times
Reputation: 7651
I think human intelligence clearly has its limits. For a long time people have not really used their potential, it was simply not necessary with farming and other simple tasks.
But nowadays I get the impression that more and more people are reaching their limits. There is a tyranny of overachievers who think everyone is or should be freaks like they are. But that is unrealistic. We are not much different from stone-age people. And there won't be any miraculous increase in intelligence.

Now would that be a good thing, we are already too intelligent, without being wise.
Reply With Quote Quick reply to this message
 
Old 05-25-2017, 10:43 PM
 
Location: Boston
1,990 posts, read 620,316 times
Reputation: 3943
Quote:
Originally Posted by thecoalman View Post
The only thing that is true AI advancement is beating us at Jeopardy. The first computer could have beat a human at chess, the issue is that it would take forever to calculate the move. Deep Blue beating Kasparov was just a demonstration of raw processing power, at the end of the day chess is just a game of math which is something computers excel at. You can't beat an opponent at chess that can calculate every possible move that are never wrong.

Jeopardy on the other hand is a completely different matter. What is intelligence? What is this? "A"

Seems simple enough but how many different fonts do you see each day some of which you have never seen before that you can immediately recognize as the letter "A". That is what separates us from a regular computer and simple things like this we are doing constantly all the time.

You can feed every conceivable font to computer and tell it that it's the letter A but if try and do that with every piece of human knowledge the task is insurmountable even if you had the ridiculous amount of raw computer power required to do it.

That is where Watson from Jeopardy comes in.You give it 30 different A's and program it to identify other A's by itself. When they ask it question it doesn't have that specific answer stored somewhere. What is important to note is Watson can be wrong. Also important to note is when it is wrong it will adjust it's programming so it's right. Scary huh?

Watson is now Dr. Watson.... they are using it in the medical field to identify abnormalities in cat scans and things of that nature way beyond what a human would be able to identify.
Chess is computable, it relies on axioms.
Computers cannot utilize consciousness, non computable problems can only be solved by humans.
Reply With Quote Quick reply to this message
 
Old 05-26-2017, 09:46 AM
 
345 posts, read 110,590 times
Reputation: 285
Quote:
Originally Posted by jonesg View Post
Chess is computable, it relies on axioms.
Computers cannot utilize consciousness, non computable problems can only be solved by humans.
What about a computer whose physical design models biological approaches?
Reply With Quote Quick reply to this message
 
Old 05-26-2017, 11:25 AM
 
Location: West Coast of Europe
19,017 posts, read 16,778,707 times
Reputation: 7651
Are there any non-computable problems in the first place?
Reply With Quote Quick reply to this message
 
Old 05-26-2017, 09:18 PM
 
34,409 posts, read 34,520,079 times
Reputation: 13474
Quote:
Originally Posted by jonesg View Post
Computers cannot utilize consciousness, non computable problems can only be solved by humans.
What is a non computable problem?
Reply With Quote Quick reply to this message
 
Old 05-27-2017, 05:48 PM
Status: "CSU Pueblo is in the NCAA D2 playoffs!" (set 4 days ago)
 
Location: Pueblo - Colorado's Second City
11,913 posts, read 19,108,504 times
Reputation: 3985
Not even close. Wait till after the technological singularity.
Reply With Quote Quick reply to this message
 
Old 05-28-2017, 10:20 AM
 
Location: midwest
1,257 posts, read 810,103 times
Reputation: 762
Probably not. Intelligence evolved under somewhat random conditions.

What could we now do to stimulate intelligence if we designed computers and software to do that?

Of course it would totally change our so called "culture" so most people would not even want to do it. So we are stuck in "Marching Morons" mode.

https://en.wikipedia.org/wiki/The_Marching_Morons
Reply With Quote Quick reply to this message
 
Old 05-29-2017, 04:52 AM
 
34,409 posts, read 34,520,079 times
Reputation: 13474
Quote:
Originally Posted by psikeyhackr View Post
What could we now do to stimulate intelligence if we designed computers and software to do that?
I'm no expert in this but my understanding is you cannot stimulate intelligence externally substantially. The IQ you have is it what it is. It can only be moderately increased a few points and even changes throughout the day.

It would have to be done with selective breeding and/or bio engineering and I don't think society is ready for that. On top of that it may be natural mechanism to prevent us from "short circuiting" for lack of a better term.

Take the eye for example, your eye is physically capable of seeing an infinite amount of information yet the average human cannot discern the difference once a frame rate hits about 120. The information between those "frames" is discarded so the brain does not have to process them. It simply fills in the blanks. This is also how many illusions are created and why film works.

Taking this a step further if you unlock the potential of the human brain you better have a very good understanding of what the fundamental differences are between Einstein and Ted Kaczynski.
Reply With Quote Quick reply to this message
 
Old 05-29-2017, 10:47 AM
 
Location: West Coast of Europe
19,017 posts, read 16,778,707 times
Reputation: 7651
I suppose there is a good reason why the brain limits its own performance, continuous overload would probably lead to serious problems. There is that old myth that only a fraction of the human brain is being used and thus there remains a huge potential to be unlocked. But I read that that notion is incorrect.

And really, aren't we intelligent enough already? The problems we just can't seem to solve had less to do with a lack of intelligence than with emotions (think of all the -isms). Even very intelligent people make stupid mistakes, not seldom thanks to emotions.
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply
Please update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:

Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Science and Technology
Similar Threads

All times are GMT -6. The time now is 04:26 AM.

2005-2017, Advameg, Inc.

City-Data.com - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 - Top