Prev: Galois group of a quintic polynomial over Q
Next: Help with a (hopefully famous) quote
From: sanboz on 16 Nov 2009 14:32

"Mark Murray" <w.h.oami(a)example.com> wrote in message
news:4b019767$0$2479$db0fefd9(a)news.zen.co.uk...
> sanboz wrote:
>> "JSH" <jstevh(a)gmail.com> wrote in message
>> news:fa5c6070-f59d-4d4a-a224-2d4e10fbf493(a)y10g2000prg.googlegroups.com...
>
> ...etc
>
> Sanboz - it is very hard to see what you wrote and what you are quoting.
>
> Please quote properly. Turning off HTML may help.
>
> M


it posts in text always

but it is Outlook Express -- it seems to forget the > marks on some posts
coming in, (about 1/4) although it is set to show them.
I haven't figured that one out

and on the longer posts it is a pain to go back and put in the >

Any Ideas?


From: Owen Jacobson on 17 Nov 2009 00:45
On 2009-11-15 22:32:52 -0500, JSH <jstevh(a)gmail.com> said:

> On Nov 15, 11:59�am, Owen Jacobson <angrybald...(a)gmail.com> wrote:
>> On 2009-11-14 23:38:49 -0500, JSH <jst...(a)gmail.com> said:
>>
>>> One of the weirder things that has emerged from my mathematical
>>> research is the possibility of continual transmission of information
>>> from the future to the past in order to CREATE the future, where key
>>> is what I call the optimal path algorithm.
>>
>>> Used against the Traveling Salesman Problem it gives you a traveler
>>> going backwards in time to meet himself, where the algorithm requires
>>> continual communication between the two travelers in order to get the
>>> optimal path.
>>
>> You have attached unwarranted, semi-mystical importance to the
>> "backwards" part of your trivial variation on bidirectional search[1].
>
> I checked the "bidirectional search" over a year ago when it was first
> trotted out as what my optimal path algorithm does.
>
> But the actual algorithm applied to TSP has a backwards Traveler, who
> is going backwards in time to meet himself along the optimal path. In
> order to calculate that path the backwards Traveler has to communicate
> with himself in the past--or you can say the Traveler moving forward
> in time has to communicate with himself in the future.
>
> It's not mystical. It's an integral part of the actual algorithm,
> where the optimal path once calculated collapses to a single traveler
> moving forward in time.

There is no "time" at all as far as any NP-complete problem is
concerned, other than the number of steps needed to reach a solution.

Your algorithm proceeds stepwise through two graphs (the "forward"
graph for your "forward" traveller, which is the original complete
weighted graph, and the "backwards" graph, which is derived from the
"forward" graph by reversing the weights, for the "backwards"
traveller). At each step, your algorithm determines which node to add
to each traveller's list of nodes next based on some vaguely-specified
criteria that is a strict function of the graph and the list of nodes
visited so far. Your algorithm terminates on a step when the lists of
notes for the two travellers intersect, and the final path is the
"forward" list up to the point where the lists intersect, followed by
the reverse of the "backwards" list.

We can model your algorithm as a perfectly ordinary recursive function
-- just like any other bidirectional search. No "communication back in
time" -- the choices made at step N of your algorithm's evaluation
depend only on the choices made in steps 0..N, not on the choices made
at some step M > N. There is nowhere any requirement that an algorithm
that searches for shortest paths through a graph start at the
beginning, and starting at other points, or at multiple points is not
innovative.

If (and this is a big if) your algorithm provably produces a
Hamiltonian cycle over the input graph such that the sum of the weights
in the circuit is minimal, then your algorithm "solves" the travelling
salesman problem. If (and this is the other big if) it provably does so
in polynomial time, it is of interest to the mathematical community.
You have proven neither thing to the satisfaction of anyone besides
yourself.

>> It does not imply time travel; you can model it just as well by
>> thinking of two travellers, one going forward through the original
>> graph, and one going forward through a reversed version of the original
>> graph.
>
> And you don't mention communication between the two travelers.

The travellers are irrelevant. You could eliminate them entirely by
phrasing your algorithm in terms of adding one node to each of two
lists of nodes at each iteration without losing any details. There is
no time, there are no travellers, there is no communication, there is
only the graph, the various structures necessary to keep track of the
algorithm's state, and the repeated application of some deterministic
function to those two things.

This is math, not amateur cosmology hour.

>> It'd serve you well to read up on how others have already solved graph
>> problems in general, and NP-complete graph problems specifically,
>> before running around trumpeting your "innovations". It's a pity you
>> won't.
>
> I have read up a lot on issues in this area. But I'm also considering
> an idea I have which seems to answer a lot of questions. From a
> purely intellectual perspective it's a fascinating algorithm to
> consider

Well, yes, there are lots of ways to refine bidirectional search for
various scenarios, just like there are lots of ways to refine any other
basic graph search. There's lots of fun to be had.

> and the implications are world shaking if true.

Not even a shiver. It's been done, it's not any faster than brute force
and no known algorithm generates solutions to the travelling salesman
problem in polynomial worst-case[0] time, including various heuristic
bidirectional search algorithms. This is undergrad-level stuff, dude.

So, to reiterate, from what little you've said this time around:

- Your algorithm is not novel. It is a variation on bidirectional search.
- You have not proven that your algorithm always generates Hamiltonian
cycles over complete weighted graphs.
- You have not proven that when your algorithm does generate a
Hamiltonian cycle the cycle will have minimal weight.
- You have not proven that it does any of this in polynomial time.

All of these things must be addressed if you want to shake the world.
The rest is your own distractions.

-o

[0] or average-case time, if you insist on solving the general problem
rather than some useful subset of the problem)

From: Mark Murray on 17 Nov 2009 02:32
sanboz wrote:
> Any Ideas?

Use something else? Thunderbird?

M
From: Jim Ferry on 17 Nov 2009 10:35
On Nov 15, 12:57 pm, "sanboz" <nos...(a)spamless.com> wrote:
> "JSH" <jst...(a)gmail.com> wrote in message
>
> news:04874ca3-d17f-4100-8439-437d1ad11d23(a)h40g2000prf.googlegroups.com...
>
> > One of the weirder things that has emerged from my mathematical
> > research is the possibility of continual transmission of information
> > from the future to the past in order to CREATE the future, where key
> > is what I call the optimal path algorithm.
>
> Join the Ashtar Command, they do that all the time.  They have "channelers"
> that get furure information from Aliens in Spaceeships waiting to land, and
> beam it to Earth where it is used for profit by the Channelers.

It's not that subatomic particles are somehow performing
a computation of optimal paths: it's just that the only
place where the probability densities of events rise
above zero is at critical points, and the only critical
points are minima, as in Feynman's explanation of QED.
As James correctly observes, these transtemporal Traveling
Salesman Problems are being "solved" all the time (at a
subatomic level), although their effects are seldom
observed macroscopically, due to mass limitations.

> > For various reasons the idea is floating around that the "end of the
> > world" is in 2012. My memory is that Sir Isaac Newton actually
> > calculated the correct year and he got 2010, but I've seen no mention
> > of that in the record, so I'm not sure why I have that number.

However, once the Higgs boson is involved, macroscopic
effects are possible. The lowest energy (therefore
most probable) effects (per unit entropy) are on human
consciousness. Thus, on the one hand we have the CERN
experiment which is continually forking the universe
into branches in which (a) the universe is destroyed
(well, rendered inhospitable to life at any rate) via
an Ice-9-like disaster, and (b) some glitch delays the
disaster. It may seem amusing or fitting somehow that
the mass-imbuing Higgs boson TSP network extended back
in time to order Sir Isaac Newton's brain so as to
understand mass (as inertia) and leave the correct
date (2010) for its own creation / world's destruction,
but the world's a funny old place.

Anyway, James has warned you. I've embellished the
warning. Now what are you going to do? Nothing?
Yeah, me neither. I feel surprisingly relaxed about
the world ending next year. Maybe that's because I'm
not insane. I'm only bringing you this message of
doom because my brain is somehow being entangled in
the Higgs boson TSP via which the future is
communicating with James Harris and Isaac Newton.
My brain and my even-now-typing fingers have no
choice in this matter. But to believe it? That
would be crazy indeed!
From: A on 17 Nov 2009 13:03
On Nov 17, 10:35 am, Jim Ferry <corkleb...(a)hotmail.com> wrote:
> On Nov 15, 12:57 pm, "sanboz" <nos...(a)spamless.com> wrote:
>
> > "JSH" <jst...(a)gmail.com> wrote in message
>
> >news:04874ca3-d17f-4100-8439-437d1ad11d23(a)h40g2000prf.googlegroups.com....
>
> > > One of the weirder things that has emerged from my mathematical
> > > research is the possibility of continual transmission of information
> > > from the future to the past in order to CREATE the future, where key
> > > is what I call the optimal path algorithm.
>
> > Join the Ashtar Command, they do that all the time.  They have "channelers"
> > that get furure information from Aliens in Spaceeships waiting to land, and
> > beam it to Earth where it is used for profit by the Channelers.
>
> It's not that subatomic particles are somehow performing
> a computation of optimal paths:  it's just that the only
> place where the probability densities of events rise
> above zero is at critical points, and the only critical
> points are minima, as in Feynman's explanation of QED.
> As James correctly observes, these transtemporal Traveling
> Salesman Problems are being "solved" all the time (at a
> subatomic level), although their effects are seldom
> observed macroscopically, due to mass limitations.
>
> > > For various reasons the idea is floating around that the "end of the
> > > world" is in 2012.  My memory is that Sir Isaac Newton actually
> > > calculated the correct year and he got 2010, but I've seen no mention
> > > of that in the record, so I'm not sure why I have that number.
>
> However, once the Higgs boson is involved, macroscopic
> effects are possible.  The lowest energy (therefore
> most probable) effects (per unit entropy) are on human
> consciousness.  Thus, on the one hand we have the CERN
> experiment which is continually forking the universe
> into branches in which (a) the universe is destroyed
> (well, rendered inhospitable to life at any rate) via
> an Ice-9-like disaster, and (b) some glitch delays the
> disaster.  It may seem amusing or fitting somehow that
> the mass-imbuing Higgs boson TSP network extended back
> in time to order Sir Isaac Newton's brain so as to
> understand mass (as inertia) and leave the correct
> date (2010) for its own creation / world's destruction,
> but the world's a funny old place.
>
> Anyway, James has warned you.  I've embellished the
> warning.  Now what are you going to do?  Nothing?
> Yeah, me neither.  I feel surprisingly relaxed about
> the world ending next year.  Maybe that's because I'm
> not insane.  I'm only bringing you this message of
> doom because my brain is somehow being entangled in
> the Higgs boson TSP via which the future is
> communicating with James Harris and Isaac Newton.
> My brain and my even-now-typing fingers have no
> choice in this matter.  But to believe it?  That
> would be crazy indeed!


James' initial warning was the first herald of the coming wave of
Trans-temporal Travelling Salesman (TTS) Consciousness. While it was
his (future-projected) genius (travelling back to his past-present
self) that first recognized the coming wave of TTS Consciousness, it
seems to be spreading rapidly--first you were "infected," and now
today I have suddenly come to understand as well that James' solution
to the Travelling Salesman Problem has opened up paths of
communication between the future universe and the present universe
which are large enough in scale that my aware-of-TTS-Consciousness
future self has informed my present-day self of the coming of TTS
Consciousness.

Since James' solution to the Travelling Salesman Problem will seem
obvious to the world as TTS Consciousness becomes widespread, don't
you think it is our duty to help announce to the world that it was
James who first discovered the solution to the Travelling Salesman
Problem, so that his name is properly recognized in history? After
all, while we now know that the future sends information back to the
past, we know very well (from history) that the past doesn't always
send all the right information forward to the future.
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: Galois group of a quintic polynomial over Q
Next: Help with a (hopefully famous) quote