cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A282166 a(n) is the minimal sum of a positive integer sequence of length n with no duplicate substrings of length greater than 1, and every number different from its neighbors.

Original entry on oeis.org

1, 3, 4, 7, 8, 12, 13, 17, 18, 22, 24, 28, 30, 35, 37, 42, 44, 49, 51, 56, 59, 64, 67, 72, 75, 81, 84, 90, 93, 99, 102, 108, 111, 117, 121, 127, 131, 137, 141, 147, 151, 158, 162, 169, 173, 180, 184, 191, 195, 202, 206, 213, 218, 225, 230, 237, 242, 249, 254, 261, 266, 274, 279, 287, 292, 300, 305, 313, 318, 326, 331, 339, 344, 352, 358, 366, 372, 380, 386, 394
Offset: 1

Views

Author

Peter Kagey, Feb 07 2017

Keywords

Comments

For example, [1,1] is not a valid sequence because 1 is self-adjacent; [1,2,3,1,2] is not valid because the substring [1,2] appears twice.

Examples

			a(1)  = 1  via [1];
a(2)  = 3  via [1,2];
a(3)  = 4  via [1,2,1];
a(4)  = 7  via [1,2,1,3];
a(5)  = 8  via [1,2,1,3,1];
a(6)  = 12 via [1,2,1,3,1,4];
a(7)  = 13 via [1,2,1,3,1,4,1];
a(8)  = 17 via [1,2,1,3,1,4,2,3];
a(9)  = 18 via [1,2,1,3,2,3,1,4,1];
a(10) = 22 via [1,2,1,3,1,4,2,3,4,1];
a(11) = 24 via [1,2,1,3,2,3,1,4,1,5,1].
		

Crossrefs

Cf. A282169 is the product analog.

Programs

  • Mathematica
    Table[Module[{s = Select[Permutations[Range@ n - 1, n], Length@ # > 1 &]}, Total@ First@ MinimalBy[#, Total] &@ DeleteCases[#, w_ /; Apply[Times, If[Length@ # > 0, Rest@ #, #] &@ Union@ Map[SequenceCount[w, #] &, s]] > 1] &@ Apply[Join, Map[MinimalBy[#, Total] &, Table[Select[Tuples[Range@ k, n], Function[w, Times @@ Boole@ {Length@ Union@ w == k, First@ #, If[n > 2, Xor @@ Rest@ #, True]} == 1 &@ Map[Length@ Split@ # == Length@ # &, {w, w[[1 ;; -1 ;; 2]], Rest[w][[1 ;; -1 ;; 2]]}]]], {k, n}]]]], {n, 7}] (* Michael De Vlieger, Mar 27 2017, Version 10 *)

Formula

For n>=4, we seem to have a(n) = a(n-1) + a(n-2) - a(n-3) + d(n), where d(n) is either 0 or 1 (with a clear formula). This observation leads to the conjecture: for n>=4, a(n) = -3/2 + 2*n + n*m/2 - m*(2*m^2+15*m+46)/24 + (-1)^n*(m%2+2)/4 + (m%2)*3/8, where m is the largest integer such that (2*m^2 + 8*m + 1 + 3*(-1)^m)/4 <= n. - Max Alekseyev, May 28 2025

Extensions

a(12)-a(21) from Lars Blomberg, Jun 10 2017
Terms a(22) onward from Max Alekseyev, Feb 04 2025