[seqfan] Re: Number of nXn (0,1)-matrices A with A^2 = J?
M. F. Hasler
seqfan at hasler.fr
Sat Mar 4 20:24:17 CET 2017
Yes, the solution mentioned in the paper is (here for n=4, m=2)
(But this is the transpose of your first solution. The transpose of your
second solution obviously works also.)
Le 4 mars 2017 12:38, "Neil Sloane" <njasloane at gmail.com> a écrit :
> Joerg, you are right. I should have stuck with
> my ugly solutions, which were
> Are there any other 4X4 0,1 matrices with A^2 = J = "all-ones" ?
> Best regards
> 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
> On Sat, Mar 4, 2017 at 3:17 AM, Joerg Arndt <arndt at jjj.de> wrote:
> > * Neil Sloane <njasloane at gmail.com> [Mar 04. 2017 09:00]:
> >> I just came across an old paper of Herb Ryser (A generalization
> >> of the matrix equation A^2=J, Linear Alg. Applic., 3 (1970),451-460)
> >> where he mentions in passing the unsolved problem of finding
> >> the number of n X n real (0,1) matrices A such that A^2 is the
> >> all-ones matrix J.
> >> He says that A must have constant row and column sums c, and c^2 = n.
> >> Also trace A = c. So n must be a square, n=c^2.
> >> There is a long entry in the Index to the OEIS under
> >> matrices, binary
> >> but this problem doesn't seem to be mentioned there.
> >> Is this sequence in the OEIS? (If so, it should
> >> be mentioned in the index entry.)
> >> I'm pretty sure I've seen the problem before, but a quick
> >> search in the OEIS didn't find it. Let n = c^2. Then there is one
> >> solution if c=1, and if someone could work out the answers for c=2 and
> >> maybe 3, that might be enough to locate it.
> >> Here is one solution for c=2:
> >> 1010
> >> 0101
> >> 1010
> >> 0101
> > This one has trace = c^2 = 4, not c as stated above.
> >> and permuting the rows is allowed - so is the answer 6 if c=2?
> >> Neil
> >> --
> >> Seqfan Mailing list - http://list.seqfan.eu/
> > --
> > Seqfan Mailing list - http://list.seqfan.eu/
> Seqfan Mailing list - http://list.seqfan.eu/
More information about the SeqFan