[ExI] Do digital computers feel?

[Jason Resch]

On Sun, Jan 1, 2017 at 7:21 PM, Rafal Smigrodzki <rafal.smigrodzki at gmail.com
> wrote:

>
>
> On Fri, Dec 30, 2016 at 9:50 AM, Jason Resch <jasonresch at gmail.com> wrote:
>
>>
>> They are possible, but up until now, all known physical laws are
>> computable,
>>
>
> ### What about the three-body problem?
>

There are solutions to this, no?
http://www.sciencemag.org/news/2013/03/physicists-discover-whopping-13-new-solutions-three-body-problem



>
> -----------------------------
>
>
>> which is why Penrose and Hameroff have to propose a speculative
>> undiscovered physics for their theory to rest upon. It is worth noting,
>> that what led Penrose to his belief that the brain does incomputable things
>> was his idea that halting problem does not apply to humans. Turning
>> discovered the Halting problem, which was the idea that no fixed computer
>> program can decide whether any other given program will complete or not.
>> But it would seem this limitation applies to humans as well. $1,000,000 was
>> offered to any person who could prove whether or not this simple program
>> ever finishes or not, the prize was never claimed:
>>
>> Step 1: Set X = 4
>> Step 2: Set R = 0
>> Step 3: For each Y from 1 to X. if both Y and (X - Y) are prime, set R = 1
>> Step 4: If R = 1, Set X = X + 2 and go to Step 2
>> Step 5: If R = 0, print X and halt
>>
>> Given that humans appear equally limited by the halting problem, the
>> entire motivation for Penrose and Hameroff to propose there exist
>> undiscovered physical laws that are incomputable evaporated.
>>
>
> ### I agree with you, and I do not say that human minds are beyond known
> physics. What I am driving at, is that digital simulations and analog
> computations are inherently different, and this can lead to surprises. No
> matter how precise the digital simulation, it will always diverge from the
> analog system.
>

True. Is it important though?


> Multiple runs of digital simulations will always run exactly the same,
> while multiple analog neural networks, no matter how closely matched in the
> beginning, will always diverge. They may diverge in predictable ways, or
> they may exhibit predictable macroscopic behavior (outputting the letter A
> in my initial example) but they will definitely differ in their microstates.
>
> This is unmapped territory. There be dragons, or maybe mice, or perhaps
> nothing at all but we don't know which one yet.
>

Indeed.

Jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20170101/5cfa1357/attachment.html>