Re: VMPC function. Question on definition of inverting

From: Bartosz Zoltak ()_at_vmpcfunction.com)
Date: 08/26/03 

Date: Tue, 26 Aug 2003 22:22:30 +0200

David Wagner
> >The simulations, carried out according to the inverting algorithm
described
> >at www.vmpcfunction.com/invert.htm , show the problem is hard for any n.
>
> Nonsense. No, it doesn't. That web page shows that one particular
> algorithm fails to work well. But there might be other, better
algorithms,
> for all you or I know.

I didn't meant to stress "SHOW" but "ANY n" rather. I was just trying to
help. I understand that what is done at the website is description of ONE
algorithm with explanations why this MIGHT be the fastest or close to the
fastest algorithm of inverting the function.

It would need to take someone with more theoretical experience and will and
belief to try to show that there is no other method, or somehow to generally
show that inverting VMPC cannot be done easily.

I don't know to what extent my algorithm of inverting VMPC would be helpful.
Having spent over 4 years analysing VMPC I got to believe that the function
cannot be esily inverted. And seeing the function's simplicity I began to
hope that maybe it would be possible to prove that it is in fact one-way.
Somehow. I definitely do not know how, but realise it would be fantastic.

I would be glad to share any of my experiences on the function to let those
who find the problem interesting continue work on it.

I was lucky to find the function and determined enough to anlyse it as well
as I could. Now someone with more theoretical knowledge, if one was willing
to, could investigate VMPC further. That I meant.

Bartosz 

--
Bartosz Zoltak
http://www.vmpcfunction.com
QPbzoltak@vmpcfunction.com
without "QP"

Relevant Pages

  • Re: Inversion of an algorithm
    ... Has anyone worked on getting *automatically* an algorithm for the ... classic eyeball test to see if a function is invertible} we will see ... That unhappy finding is that for any y value between 1.0 and ^(1/ ... There is logical problem inverting a function that is not one-to-one ...
    (comp.programming)
  • Re: Inversion of an algorithm
    ... Has anyone worked on getting *automatically* an algorithm for the ... inverse function from B to A? ... That unhappy finding is that for any y value between 1.0 and ^(1/ ... There is logical problem inverting a function that is not one-to-one ...
    (comp.programming)
  • Re: bilinear pairing between special groups
    ... for inverting the map ... efficient algorithm for finding the cyclic group of order q sitting ... This is not the discrete logarithm problem, since we do now know the ... could be done efficiently, but what happens with the bilinear pairing, ...
    (sci.math)
  • Square-root Kalman Filter Algorithm
    ... Currently I am using traditional Kalman Filter algorithm for my problem ... the code has problems with inverting some matrices. ... square-root algoritm is more stable. ...
    (comp.dsp)
Quantcast