login

Revision History for A087811

(Bold, blue-underlined text is an addition; faded, red-underlined text is a deletion.)

Showing entries 1-10 | older changes
Numbers k such that ceiling(sqrt(k)) divides k.
(history; published version)
#187 by Michael De Vlieger at Sun Sep 29 13:18:15 EDT 2024
STATUS

reviewed

approved

#186 by Peter Munn at Sun Sep 29 12:45:28 EDT 2024
STATUS

proposed

reviewed

#185 by Miquel A. Fiol at Sun Sep 29 12:05:06 EDT 2024
STATUS

editing

proposed

Discussion
Sun Sep 29
12:45
Peter Munn: #12:01: your "when possible" route forward sounds like you understood.
#184 by Miquel A. Fiol at Sun Sep 29 12:04:48 EDT 2024
FORMULA

From Miquel A. Fiol, Sep 28 2024 (Start):

a(n) = Sum_{j=0..floor(n/2)} binomial(c+n-2*j-1,n-2*j) or, alternatively,

a(n) = Sum_{j=0..n}(-1)^(n-j)*binomial(c-j,j), in both cases, with c=2 and the shift n -> n-1.

For other values of c we obtain: A002623 (c=3); A001752 (c=4); A001753 (c=5); A001769 (c=6); A001779 (c=7), A001780 (c=8); A001781 (c=9); A001786 (c=10), A001808 (c=11).

(End)

#183 by Kevin Ryde at Sat Sep 28 20:35:06 EDT 2024
STATUS

proposed

editing

Discussion
Sun Sep 29
04:11
Peter Munn: @Miquel. Your full project here (with the table of sequences you showed earlier) looks interesting, but we should display it on a suitable stage. I'll come back in an hour or two and summarise some options.
05:20
Miquel A. Fiol: OK Peter, I'll wait for your suggestions. Thanks
05:45
Miquel A. Fiol: @Kevin. Sorry about that. With n->n-1 I intended to mean that for the case c=2 we should write a(n) = Sum_{j=0..floor(n/2)} binomial(c+(n-1)-2*j-1,(n-1)-2*j) . This is because the offset of A087811 is 1 instead of 0 (which is the offset for the other mentioned sequences A02623 (c=3), A001752 (c=4), etc.
05:54
Miquel A. Fiol: @Andrew. I'm waiting the intructions from some Editor about how to proceed with my contribution here.
09:14
Peter Munn: As Andrew says, your formula proposed here would be much more appropriately placed in the table sequence A059260.
  (Thanks to Andrew for finding (the rather old) A059260 is my suggested "new table sequence" from my comment on revision #167.)

Similarly, the formula that you recently added to A005993 is much more appropriately placed in the table A062135. A search for [A008619 A005993 A005995 A018211] finds this sequence.

However, in revision #167 here you showed a formula linking all these and more sequences, thereby including and linking the tables A059260 and A062135 in a 3-dimensional table indexed by n, c1 and c2. This seems to be the significant novel aspect of your work. So how to exhibit it well?

Andrew mentioned earlier that there are difficult issues with creating a sequence that represents a 3D table, and I also caution against that route. So ...

Option 1.
Take the route that could be used by someone who has not joined our community, and author a paper or web page about your findings. You could still reference all the OEIS A-numbers.
  Also a paper, however short, could be uploaded to be stored with a sequence, such as A062135. https://oeis.org/SubmitB.html ends with a section on uploading other files associated with a sequence.

Option 2.
Similar to the web page variant of option 1. As Kevin mentioned earlier, your findings could be put on the OEIS wiki, with links from the wiki page to the sequences and vice versa. (Kevin, do you have any suggested wiki pages Miquel could look at for inspiration?)

Option 3.
Choose one of the mentioned table sequences, and include in that sequence the full "3D" formula from revision #167 here. My preference would be to refer readers from your 3D formula in the formula section of that sequence to the example section, where you could put your full array of sequences from revision #167 here. This array of sequences should appear in one place only. (E.g., if it is in A062135, refer readers to it from A059260.) Then there is no problem of improvements not being made to all copies.

In any event, your formula contribution here can be removed and instead refer readers to A059260 for readers to see how this sequence relates to A002623, A001752 etc. (without any mention here of parameter c). You could do this in the xrefs and this can happen now as A059260 already contains some relevant information.
12:01
Miquel A. Fiol: OK, Peter. Thank you very much for detailing me the possible options.
I'll choose Option 3. Thus, I'll removed my contribution here, leaving only the xref A059260.
For the moment, I cannot do anything more because I already have three current submissions.
When possible, I'll put the full "3D" formula in  in the formula section, together with the array of sequences in in the example section (if I undestood well).
#182 by Miquel A. Fiol at Sat Sep 28 15:56:01 EDT 2024
STATUS

editing

proposed

Discussion
Sat Sep 28
16:50
Andrew Howroyd: but you have left it?
20:35
Kevin Ryde: This is a revert?  If "shift n -> n-1" means what it says then definitely no on that.  a(n) = something must mean a(n) = something.  Anything else is too confusing.
#181 by Miquel A. Fiol at Sat Sep 28 15:52:17 EDT 2024
LINKS

Vincenzo Librandi, <a href="/A087811/b087811.txt">Table of n, a(n) for n = 1..300</a>

Discussion
Sat Sep 28
15:55
Miquel A. Fiol: I see, better to withdraw my entries here.
#180 by Miquel A. Fiol at Sat Sep 28 15:51:07 EDT 2024
LINKS

Vincenzo Librandi, <a href="/A087811/b087811.txt">Table of n, a(n) for n = 1..300</a>

R. H. Hammack and G. D. Smith, <a href="https://doi.org/10.26493/1855-3974.856.4d2">Cycle bases of reduced powers of graphs</a>, Ars Math. Contemp. 12 (2017) 183-203.

#179 by Miquel A. Fiol at Sat Sep 28 15:48:13 EDT 2024
COMMENTS

Conjecture: a(n) is the independence number of the n-supertoken (or reduced n-th power) FF_n(S_c) of the star graph S_c with c(>=3) edges. - Miquel A. Fiol, Sep 28 2024

STATUS

proposed

editing

#178 by Miquel A. Fiol at Sat Sep 28 15:36:43 EDT 2024
STATUS

editing

proposed