[seqfan] Re: Modular Partitions
Roland Bacher
Roland.Bacher at ujf-grenoble.fr
Thu May 1 16:21:47 CEST 2014
The link with Molien series is easy to explain:
Molien series count dimensions of polynomials invariant
under the action of a linear group.
Cyclic groups act diagonally by multiplication with a root of 1
(of order divisible by N for Z/NZ).
Consider now the action of Z/NZ on R[x_0,..x_{N-1}]
acting by multiplication with $e^{2ij\pi/N} on x_j.
The dimension of invariant polynomials of homogeneous degree
k is then exactly the modular partition number n,k
mentionned by Jens Voss
Molien series are rational and can be computed using a determinantal
formula. This should allow to prove the observed symetry.
Roland
On Wed, Apr 30, 2014 at 04:03:46PM -0400, Neil Sloane wrote:
> A brief look at the triangle suggests that row n gives
> the coeffts in the Molien series for the cyclic group of order n.
> See A007997, A008610, A008646, A032191.
> If true this should follow immediately from the definition.
> A good reference is B. Sturmfels, Algorithms in Invariant Theory, Springer,
> '93, p. 65.
>
> I don't have time to look into this further right now.
>
> Neil
>
>
>
> On Wed, Apr 30, 2014 at 3:19 PM, Jens Voß <jens at voss-ahrensburg.de> wrote:
>
> >
> > Hi there, sequence fans,
> >
> > I was playing around with what I call "modular partition numbers":
> > Essentially different ways to write the neutral element of the group Z/nZ
> > as a sum of length k (for given n, k > 0).
> >
> > For example, for n = 5 and k = 4, we have thepartitions
> >
> > 0+0+0+0 = 0
> > 0+0+1+4 = 5 = 0
> > 0+0+2+3 = 5 = 0
> > 0+1+1+3 = 5 = 0
> > 0+1+2+2 = 5 = 0
> > 0+2+4+4 = 10 = 0
> > 0+3+3+4 = 10 = 0
> > 1+2+3+4 = 10 = 0
> > 1+3+3+3 = 10 = 0
> > 3+4+4+4 = 15 = 0
> >
> > so the number of 5-modular partitions of length 4 is 10.
> >
> > I computed the the values for n + k < 20 (as a square array read by
> > antidiagonals), and was somewhat surprised that this sequence isn't yet in
> > the database (even though several of the rows resp. columns are). However,
> > I was even more surprised to find that the array is symmetric in n and k:
> >
> > 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
> > 1 1 1
> > 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9
> > 9 10
> > 1 2 4 5 7 10 12 15 19 22 26 31 35 40 46 51
> > 57
> > 1 3 5 10 14 22 30 43 55 73 91 116 140 172 204 245
> > 1 3 7 14 26 42 66 99 143 201 273 364 476 612 776
> > 1 4 10 22 42 80 132 217 335 504 728 1038 1428 1944
> > 1 4 12 30 66 132 246 429 715 1144 1768 2652 3876
> > 1 5 15 43 99 217 429 810 1430 2438 3978 6310
> > 1 5 19 55 143 335 715 1430 2704 4862 8398
> > 1 6 22 73 201 504 1144 2438 4862 9252
> > 1 6 26 91 273 728 1768 3978 8398
> > 1 7 31 116 364 1038 2652 6310
> > 1 7 35 140 476 1428 3876
> > 1 8 40 172 612 1944
> > 1 8 46 204 776
> > 1 9 51 245
> > 1 9 57
> > 1 10
> > 1
> >
> > I haven't been able to come up with a formula for the numbers (neither
> > recursive nor direct), and I don't see an immediate reason for the symmetry
> > either (some sort of dualism). Can somebody find a formula or explain why
> > the array is symmetric?
> >
> > Best regards,
> > Jens
> >
> > _______________________________________________
> >
> > Seqfan Mailing list - http://list.seqfan.eu/
> >
>
>
>
> --
> Dear Friends, I have now retired from AT&T. New coordinates:
>
> Neil J. A. Sloane, President, OEIS Foundation
> 11 South Adelaide Avenue, Highland Park, NJ 08904, USA.
> Also Visiting Scientist, Math. Dept., Rutgers University, Piscataway, NJ.
> Phone: 732 828 6098; home page: http://NeilSloane.com
> Email: njasloane at gmail.com
>
> _______________________________________________
>
> Seqfan Mailing list - http://list.seqfan.eu/
More information about the SeqFan
mailing list