Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-12 Thread 'spudboy...@aol.com' via Everything List 
 Isn't it fair to state that we will measure its intelligence, by what it 
actually accomplishes? Things that impact us humans.
On Sunday, June 11, 2023 at 07:14:31 PM EDT, Tomasz Rola  
wrote:  
 
 On Sun, Jun 11, 2023 at 02:52:01PM -0700, Brent Meeker wrote:
> The next level will be when an AI makes /*itself*/ smarter.
> 
> Brent

But, but, how are humans going to measure its smartness (to know, for
one example, that it is not faking its own progress).

-- 
Regards,
Tomasz Rola

--
** A C programmer asked whether computer had Buddha's nature.      **
** As the answer, master did "rm -rif" on the programmer's home    **
** directory. And then the C programmer became enlightened...      **
**                                                                **
** Tomasz Rola          mailto:tomasz_r...@bigfoot.com            **

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/ZIZVUtS0s2WtftUa%40tau1.ceti.pl.
  

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/599286656.1037647.1686596547401%40mail.yahoo.com.

Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-12 Thread 'spudboy...@aol.com' via Everything List 
 Moreover, as an engineer, how would you apply this and to accomplish what? 
On Monday, June 12, 2023 at 07:41:57 AM EDT, John Clark 
 wrote:  
 
 On Sun, Jun 11, 2023 at 7:14 PM Tomasz Rola  wrote:


> But, but, how are humans going to measure its smartness 

By using a stopwatch. An Artificial Intelligence has recently found a machine 
language sorting program that is smaller and faster than anything any human 
being has seen before. And a Artificial Intelligence has also found a 
fundamentally faster way to multiply matrices, and nearly every large program 
needs to multiply matrices at some point, especially if the program involves 
graphics or, you guessed it, Artificial Intelligence. To me that certainly 
sounds like the sign post of an exponential process in the making, and 
exponential processes tend to end in a singularity. 
AI Reveals New Possibilities in Matrix Multiplication

to know, for one example, that it is not faking its own progress

In this case it would take more brain power to successfully fake its progress 
then it would take to actually make real progress.  John K Clark    See what's 
on my new list at  Extropolis
mnp




-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/CAJPayv36_Ubx0-O5WpkjuCEA5X3wZA07_drmAnX-D3b%2BEOqO8A%40mail.gmail.com.
  

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/644193885.1715598.1686596390340%40mail.yahoo.com.

Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-12 Thread John Clark 
On Sun, Jun 11, 2023 at 7:14 PM Tomasz Rola  wrote:

*> But, but, how are humans going to measure its smartness *


By using a stopwatch. An Artificial Intelligence has recently found a
machine language sorting program that is smaller and faster than anything
any human being has seen before. And a Artificial Intelligence has also
found a fundamentally faster way to multiply matrices, and nearly every
large program needs to multiply matrices at some point, especially if the
program involves graphics or, you guessed it, Artificial Intelligence. To
me that certainly sounds like the sign post of an exponential process in
the making, and exponential processes tend to end in a singularity.

AI Reveals New Possibilities in Matrix Multiplication


*to know, for one example, that it is not faking its own progress*


In this case it would take more brain power to successfully fake its
progress then it would take to actually make real progress.
John K ClarkSee what's on my new list at  Extropolis

mnp

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/CAJPayv36_Ubx0-O5WpkjuCEA5X3wZA07_drmAnX-D3b%2BEOqO8A%40mail.gmail.com.

Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-11 Thread Russell Standish 
I haven't read the original article, but given the timing, I believe
this was the application of AI in a competitive environment to
optimise sorting algorithms of short sequences (<8 items, IIUC).

Which reminds of work done by Danny Hillis (of Thinking Machines fame)
in the late 1980s who set up a coevolutionary genetic algorithm to
evolve sorting algorithms that beat classic sorting algorithms. ISTM -
they've just rediscovered his work.

Danny has a paper in the ALifeII proceedings describing that work.

On Sun, Jun 11, 2023 at 02:52:01PM -0700, Brent Meeker wrote:
> The next level will be when an AI makes itself smarter.
> 
> Brent
> 
> On 6/11/2023 7:34 AM, John Clark wrote:
> 
> If this is not an example of an AI helping to develop a smarter AI then I
> don't know what is. People are gonna be having a very hard time trying to
> convince themselves this is just a glorified autocomplete program. 
> 
>     AlphaDev - DeepMind AI Discovers Better Algorithms for Foundational
> Computing
> 
> John K Clark    See what's on my new list at  Extropolis
> pui
> --
> You received this message because you are subscribed to the Google Groups
> "Everything List" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to everything-list+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/
> everything-list/
> CAJPayv3Tn98n4yZL61z0Ry__BNvf9WJ25VZOXrrtLAn_fYt0FQ%40mail.gmail.com.
> 
> 
> --
> You received this message because you are subscribed to the Google Groups
> "Everything List" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email
> to everything-list+unsubscr...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/
> everything-list/46449833-251f-cdee-5129-84a7e83f17c5%40gmail.com.

-- 


Dr Russell StandishPhone 0425 253119 (mobile)
Principal, High Performance Coders hpco...@hpcoders.com.au
  http://www.hpcoders.com.au


-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/20230611231447.GA15947%40zen.

Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-11 Thread Tomasz Rola 
On Sun, Jun 11, 2023 at 02:52:01PM -0700, Brent Meeker wrote:
> The next level will be when an AI makes /*itself*/ smarter.
> 
> Brent

But, but, how are humans going to measure its smartness (to know, for
one example, that it is not faking its own progress).

-- 
Regards,
Tomasz Rola

--
** A C programmer asked whether computer had Buddha's nature.  **
** As the answer, master did "rm -rif" on the programmer's home**
** directory. And then the C programmer became enlightened...  **
** **
** Tomasz Rola  mailto:tomasz_r...@bigfoot.com **

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/ZIZVUtS0s2WtftUa%40tau1.ceti.pl.

Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-11 Thread Brent Meeker 

The next level will be when an AI makes /*itself*/ smarter.

Brent

On 6/11/2023 7:34 AM, John Clark wrote:
If this is not an example of an AI helping to develop a smarter AI 
then I don't know what is. People are gonna be having a very hard time 
trying to convince themselves this is just a glorified autocomplete 
program.


AlphaDev - DeepMind AI Discovers Better Algorithms for Foundational 
Computing <https://www.youtube.com/watch?v=6rfh72KbMug>


John K Clark    See what's on my new list at Extropolis 
<https://groups.google.com/g/extropolis>

pui
--
You received this message because you are subscribed to the Google 
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send 
an email to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/CAJPayv3Tn98n4yZL61z0Ry__BNvf9WJ25VZOXrrtLAn_fYt0FQ%40mail.gmail.com 
<https://groups.google.com/d/msgid/everything-list/CAJPayv3Tn98n4yZL61z0Ry__BNvf9WJ25VZOXrrtLAn_fYt0FQ%40mail.gmail.com?utm_medium=email_source=footer>.


--
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/46449833-251f-cdee-5129-84a7e83f17c5%40gmail.com.

Re: DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-11 Thread 'spudboy...@aol.com' via Everything List 
 No objection to this at all, as long as it can make new medical discoveries 
that lead to new cures, innovations in energy that we humans missed, and the 
lot. If AI cannot help our species, I say its worse than dangerous, it's 
useless to us!
On Sunday, June 11, 2023 at 10:34:57 AM EDT, John Clark 
 wrote:  
 
 If this is not an example of an AI helping to develop a smarter AI then I 
don't know what is. People are gonna be having a very hard time trying to 
convince themselves this is just a glorified autocomplete program. 
AlphaDev - DeepMind AI Discovers Better Algorithms for Foundational Computing

John K Clark    See what's on my new list at  Extropolis
pui

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/CAJPayv3Tn98n4yZL61z0Ry__BNvf9WJ25VZOXrrtLAn_fYt0FQ%40mail.gmail.com.
  

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/795909669.1446783.1686507004790%40mail.yahoo.com.

DeepMind AI Discovers Better Algorithms for Foundational Computing

2023-06-11 Thread John Clark 
If this is not an example of an AI helping to develop a smarter AI then I
don't know what is. People are gonna be having a very hard time trying to
convince themselves this is just a glorified autocomplete program.

AlphaDev - DeepMind AI Discovers Better Algorithms for Foundational
Computing <https://www.youtube.com/watch?v=6rfh72KbMug>

John K ClarkSee what's on my new list at  Extropolis
<https://groups.google.com/g/extropolis>
pui

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/everything-list/CAJPayv3Tn98n4yZL61z0Ry__BNvf9WJ25VZOXrrtLAn_fYt0FQ%40mail.gmail.com.