login

Revision History for A331800

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

Showing entries 1-10 | older changes
a(1) = 1; thereafter a(n) is the smallest number > a(n-1) which is neither of the form 2*a(i) nor Sum_{j=1..n-1} ( b_j*a(j) ) where 0 < i < n, b_j = 0 or 1.
(history; published version)
#50 by N. J. A. Sloane at Sat Feb 29 11:27:28 EST 2020
STATUS

editing

approved

#49 by N. J. A. Sloane at Sat Feb 29 11:27:06 EST 2020
NAME

a(1) = 1; thereafter a(n) is the next smallest number after > a(n-1) which cannot be represented in is neither of the form 2*a(i) and nor Sum_{j=1..n-1} ( b_j*a(j) ) where 0 < i < n, b_j = 1 0 or 0. a(1) = 1. Inserting the additional term a(0) = 2 results in a so-called complete sequence after sorting.

DATA

2, 1, 3, 5, 7, 17, 19, 50, 64, 152, 190, 470, 598, 1448, 1828, 4472, 5668, 13796, 17452, 42584, 53920, 131408, 166312, 405560, 513400, 1251584, 1584208, 3862592, 4889392, 11920400, 15088816, 36788000, 46566784, 113532416, 143710048, 350376032, 443509600, 1081305728

OFFSET

0,1

1,2

COMMENTS

Inserting the additional term a(0) = 2 would result in a so-called complete sequence after sorting. (The sorted sequence, from smallest to largest, will then meet Brown's criterion.)

STATUS

proposed

editing

Discussion
Sat Feb 29
11:27
N. J. A. Sloane: Edited
#48 by Hugo Pfoertner at Thu Jan 30 13:25:48 EST 2020
STATUS

editing

proposed

Discussion
Thu Feb 27
18:18
M. F. Hasler: In NAME, (1) maybe clarify "next" => "next larger"? (2) what means "represented in the form X and Y"? Could you please rephrase this? (3) I would suggest moving "Inserting the additional term a(0) = 2 results in a so-called complete sequence after sorting." to COMMENTS and just leave a(0)=2 in NAME.
#47 by Hugo Pfoertner at Thu Jan 30 13:25:01 EST 2020
KEYWORD

nonn,more,changed

STATUS

proposed

editing

Discussion
Thu Jan 30
13:25
Hugo Pfoertner: "more" no longer needed.
#46 by Hugo Pfoertner at Tue Jan 28 13:10:20 EST 2020
STATUS

editing

proposed

#45 by Hugo Pfoertner at Tue Jan 28 13:09:04 EST 2020
NAME

a(n) is the next number after a(n-1) which cannot be represented in the form 2*a(i) and Sum_{j=1..n-1} ( b_j*a(j) ) where 0 < i < n, b_j = 1 or 0. The a(1) = 1. Inserting the additional term a(0) = 2 results in a so-called complete sequence starts: a(0) = 2; a(1) = 1after sorting.

DATA

2, 1, 3, 5, 7, 17, 19, 50, 64, 152, 190, 470, 598, 1448, 1828, 4472, 5668, 13796, 17452, 42584, 53920, 131408, 166312, 405560, 513400, 1251584, 1584208, 3862592, 4889392, 11920400, 15088816, 36788000, 46566784, 113532416, 143710048, 350376032, 443509600, 1081305728

COMMENTS

This sequence is a complete sequence.

EXTENSIONS

More terms, using Rémy Sigrist's C++ at A331811 from Hugo Pfoertner, Jan 28 2020

STATUS

proposed

editing

#44 by Zhandos Mambetaliyev at Tue Jan 28 10:48:22 EST 2020
STATUS

editing

proposed

Discussion
Tue Jan 28
11:01
Zhandos Mambetaliyev: The comment in the PARI code says about generating from the second element of the sequence.  After all, there are many other sequences where the first elements of the sequence are explicitly set, I do not see a problem.
11:03
Zhandos Mambetaliyev: ...whis the missing term 2 ...
11:53
Hugo Pfoertner: I disagree. With other sequences, however, the starting terms are e.g. be set to start a recursion. In the present case, the first term a(0)=2 is in the wrong position. In order to arrive at a sequence that fulfills the condition of completeness, it would have to be between 1 and 3. I mean that terms that you would need in a different position to fulfill an additional condition cannot simply be put at the beginning.
12:22
Zhandos Mambetaliyev: Okay, You can edit as you see fit.
#43 by Zhandos Mambetaliyev at Tue Jan 28 10:48:16 EST 2020
NAME

a(0) = 2; a(1) = 1; a(n) = is the next number after a(n-1) which cannot be represented in the form 2*a(i) and Sum_{j=1..n-1} ( b_j*a(j) ) where 0 < i < n, b_j = 1 or 0. The sequence starts: a(0) = 2; a(1) = 1.

PROG

upto(lim)={my(a=[1], b=[]); for(i=1, lim, forsubset(#a, x, b=concat(b, [vecsum(vecextract(a, x))])); b=setminus(vecsort(b, , 8), a); for(j=1, #a, b=concat(b, [2*a[j]]); b=vecsort(b, , 8)); if(setsearch(b, i)==0, a=concat(a, [i]); a=vecsort(a, , 8)) ); a}

STATUS

proposed

editing

#42 by Zhandos Mambetaliyev at Tue Jan 28 10:31:43 EST 2020
STATUS

editing

proposed

Discussion
Tue Jan 28
10:43
Hugo Pfoertner: As far as I understand, the initial "2" is needed to satisfy the "complete sequence" criterion given in the corresponding WOM link. In my opinion, this is very artificial. I would discard the term a(0) and instead say in the comment something like "apart from the missing term " this is a complete sequence".
10:45
Hugo Pfoertner: ...missing term 2 ...
#41 by Zhandos Mambetaliyev at Tue Jan 28 10:31:36 EST 2020
PROG

(PARI) /* a(n) for n>0 */

(PARI) upto(lim)={my(a=[1], b=[]); for(i=1, lim, forsubset(#a, x, b=concat(b, [vecsum(vecextract(a, x))])); b=setminus(vecsort(b, , 8), a); for(j=1, #a, b=concat(b, [2*a[j]]); b=vecsort(b, , 8)); if(setsearch(b, i)==0, a=concat(a, [i]); a=vecsort(a, , 8)) ); a}

STATUS

proposed

editing