On 2013-12-14, Peter Mueller <[email protected]> wrote:
> The function delsarte_bound_hamming_space(n, d, q, isinteger=False, 
> return_data=False, solver='PPL') offers the option isinteger=True. As the 
> doc doesn't tell it, I got a little curious what is assumed to be integral. 
> Looking at the implementation it turns out that the distance distribution 
> is assumed to consist of integers. However, for non-linear codes these 
> numbers rarely are integral!
>
> Checking the bounds obtained by this didn't produce anything which 
> contradicts known lower bounds, but it improves quite a few known upper 
> bounds in Agrell's and Brouwer's tables (modulo the fact that the MIP 
> solvers are based on floating point LP solvers and thus don't give proven 
> results.)
>
> So I seriously doubt that the isinteger=True is based on a valid 
> mathematical theorem, or is there some extension of Delsarte's Theorem 
> which allows to assume that the distance distribution in an optimal code 
> consists of integers?

No, certainly no such theorem is known. (Lex Scrijver has invented
stronger bounds, where one fixes a word, and then one deals with
weight distributions relative to this word - but they are no longer LP
bounds, but SDP bounds).

It's a pure carelessness on my side that led to this bug in the Sage
code. Thanks for your report!  I've opened a ticket for it:
http://trac.sagemath.org/ticket/15520

Best,
Dima


>
> -- Peter Mueller
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to