On Sat, 4 Sep 2021 17:52:27 -0500
olcott <NoOne@NoWhere.com> wrote:
On 9/4/2021 5:39 PM, Richard Damon wrote:
On 9/4/21 6:15 PM, olcott wrote:
On 9/4/2021 5:01 PM, Richard Damon wrote:
On 9/4/21 4:59 PM, olcott wrote:
On 9/4/2021 3:48 PM, Richard Damon wrote:
On 9/4/21 2:47 PM, olcott wrote:
On 9/4/2021 1:15 PM, Richard Damon wrote:
On 9/4/21 1:46 PM, olcott wrote:
On 9/4/2021 12:34 PM, Richard Damon wrote:
On 9/4/21 1:21 PM, olcott wrote:
On 9/4/2021 12:13 PM, Richard Damon wrote:
He says:
If M enters an infinite loop, then no matter how long we >>>>>>>>>>>>> wait, we can
never be sure that M is in fact in a loop. It may simply >>>>>>>>>>>>> be a case
of a
very long computation. What we need is an algorithm that >>>>>>>>>>>>> can determine
the correct answer for any M and w by performing some >>>>>>>>>>>>> analysis on the
machine's description and the input. But as we now show, >>>>>>>>>>>>> no such algorithm exists.
Thus he recognized that the issue with a simulating
decider would be it
No he recognized the very obvious issue of using a
simulator instead of
a decider. No one besides me has ever considered a
simulating halt decider that examines the simulated
execution trace for non halting
patterns of behavior.
Nope, He understood the issues involved. Maybe if you had >>>>>>>>>>> studied some
of the field you would know that the limitation of Halt
Deciding by Simulating are WELL known, and have been shown >>>>>>>>>>> to be impossible in general.
In the text that you referenced he was only referring to
using a simulator as a decider. He was not referring to
using a simulating decider that examines the execution trace >>>>>>>>>> of the simulation to look for
non halting behavior patterns.
Nope, maybe he doesn't explicitly call it that, but his words >>>>>>>>> definitely
reference the well known and studied limitation of simulation >>>>>>>>> for halt
deciding.
Of course. If you want to tell if an infinite loops halts you
sure as Hell can't simply wait and see what happens.
It is getting to the point where I am convinced that you are
simply lying. If you are aware of any source besides me that
proposes a simulating halt decider that specifically examines
the execution trace of its simulation to match non-halting
behavior patterns of its input then PUT UP OR SHUT UP !!!
Most of the stuff I know was pre-internet, so not easy to find.
Here is one example of a reference to this from a decade ago:
https://math.stackexchange.com/questions/27606/detecting-cycles-in-off-line-turing-machines
This mentions one of the techniques used for detecting SOME
forms of infinite loops.
Here is another person needing to solve the halting problem for
a limited case, and was given a few examples of classical
methods (like detecting repeating state) to detect an infinite
loop.
https://try2explore.com/questions/10671161
And then there is this article on detecting the non-termination
of Turing Machines, to look for solutions to things like the
Busy-Beaver problem:
https://dl.acm.org/doi/pdf/10.5555/1273694.1273703
While not specifically a 'simulating Halt Decider' it is trying
to solve
the same basic problem.
Maybe the fact that you refuse to study the field means you
don't recognize that, and are dooming yourself to repeating
all the mistakes that have been worked through over the
century,
PUT UP OR SHUT UP !!!
PUT UP OR SHUT UP !!!
PUT UP OR SHUT UP !!!
PUT UP OR SHUT UP !!!
Will you now SHUT UP that NO ONE has looked at this before?
My original words included to the same extent that I have.
None-the-less is seems clear that you now do understand that
when Linz referred to a UTM he was only referring to using a UTM
as a halt decider, not using a hybrid UTM halt decider that
examines the execution trace of its input.
Nope, because I remember when I was in school, it was already
established that Simulating Halt Deciding did not show much
promise as there were serious limits as to what you could detect.
Linz knew that and knew that mentiones in passing that it
couldn't know enough to make the decision.
Also, since he proved it for ALL Halt deciders, he proved it for
Simulating Halt Deciders, as those are within the class of Halt
Deciders, and can't do anything that a 'generic' Halt Decider
can't do.
None-the-less int main() { H1(P,P); } does correctly report that
its input halts on the basis that H(P,P) does correctly report
that its input never halts.
But since H^ was built on H, it is H that needs to get the answer
right, not H1, and it doesn't
If you want to claim that they are the same machine, you need to
explain how they give different answers for the same input, which
shows they are Computations.
If you knew the x86 language and software engineering well enough
you would know that the following execution trace of the
simulation of P(P) matches the infinite recursion behavior pattern
and you would know that the infinite recursion behavior pattern is
correct.
Nope, since it skips over the CONDITIONAL code of H.
That code needs to be traced and shown to be unconditional.
THAT YOU SIMPLY DON'T KNOW THESE THINGS WELL ENOUGH IS PROVEN BY
THE FACT THAT YOU ALWAYS CHANGE THE SUBJECT INSTEAD OF DIRECT
POINTING OUT ANY ERROR
WRONG. I keep pointing out that you build your arguement on false
foundations.
>
Begin Local Halt Decider Simulation at Machine Address:c36
[00000c36][002117ca][002117ce] 55 push ebp
[00000c37][002117ca][002117ce] 8bec mov ebp,esp
[00000c39][002117ca][002117ce] 8b4508 mov eax,[ebp+08]
[00000c3c][002117c6][00000c36] 50 push eax // push P
[00000c3d][002117c6][00000c36] 8b4d08 mov ecx,[ebp+08]
[00000c40][002117c2][00000c36] 51 push ecx // push P
[00000c41][002117be][00000c46] e820fdffff call 00000966 // call
H(P,P)
[00000c36][0025c1f2][0025c1f6] 55 push ebp
[00000c37][0025c1f2][0025c1f6] 8bec mov ebp,esp
[00000c39][0025c1f2][0025c1f6] 8b4508 mov eax,[ebp+08]
[00000c3c][0025c1ee][00000c36] 50 push eax // push P
[00000c3d][0025c1ee][00000c36] 8b4d08 mov ecx,[ebp+08]
[00000c40][0025c1ea][00000c36] 51 push ecx // push P
[00000c41][0025c1e6][00000c46] e820fdffff call 00000966 // call
H(P,P) Local Halt Decider: Infinite Recursion Detected Simulation
Stopped
This infinite recursion detection criteria are met by the above
execution trace:
(a) P calls H twice in sequence from the same machine address.
(b) With the same parameters: (P,P) to H.
(c) With no conditional branch or indexed jump instructions in the
execution trace of P.
Only because the trace is incorrect.
(d) We know that there are no return instructions in H because we
know that H is in pure simulation mode.
The H can NEVER answer even as a top level machine, so THAT is
false too.
Remember there is no such thing a 'Pure Simulator Mode', something
is or it isn't a Pure Simulator. H isn't if it ever answer H(H^,H^)
That the entire time that the halt decider is making its halt status
decision the halt decider has no behavior what-so-ever that can have
any effect on the behavior of its simulated input seems to be beyond
your intellectual capacity to comprehend.
The ad hominem attack is a logical fallacy: attack the argument and not
the person and progress might be made.
/Flibble
Sysop: | Keyop |
---|---|
Location: | Huddersfield, West Yorkshire, UK |
Users: | 365 |
Nodes: | 16 (2 / 14) |
Uptime: | 64:15:49 |
Calls: | 7,774 |
Files: | 12,908 |
Messages: | 5,749,824 |