login

Reminder: The OEIS is hiring a new managing editor, and the application deadline is January 26.

Starting with input 0, find the binary value of the input. Then interpret resulting string of 1's and 0's as prime-based numbers, as follows: 0's are separators, uninterrupted strings of 1's are interpreted from right to left as exponents of the prime numbers. Output is returned as input for the next number in sequence.
4

%I #44 Apr 05 2020 21:19:07

%S 0,1,2,3,4,5,6,9,10,15,16,11,12,25,50,147,220,6125,1968750,

%T 89142864525,84252896510182189218,

%U 34892570216750728458698250328871491829901861750593684043

%N Starting with input 0, find the binary value of the input. Then interpret resulting string of 1's and 0's as prime-based numbers, as follows: 0's are separators, uninterrupted strings of 1's are interpreted from right to left as exponents of the prime numbers. Output is returned as input for the next number in sequence.

%C From _Antti Karttunen_, Oct 15 2016: (Start)

%C Iterates of map f : n -> A005940(1+n), (Doudna-sequence, but with starting offset zero) starting from the initial value 0. Conversely, the unique infinite sequence such that a(n) = A156552(a(n+1)) and a(0) = 0.

%C Note that map f can also form cycles, like 7 <-> 8 (A005940(1+7) = 8, A005940(1+8) = 7).

%C On the other hand, this sequence cannot ever fall into a loop because 0 is not in the range of map f, for n=0.., while f is injective on [1..]. Thus the values obtained by this sequence are not bounded, although there might be more nonmonotonic positions like for example there is from a(10) = 16 to a(11) = 11.

%C The formula A008966(a(n+1)) = A085357(a(n)) tells that the squarefreeness of the next term a(n+1) is determined by whether the previous term a(n) is a fibbinary number (A003714) or not. Numerous other such correspondences hold, and they hold also for any other trajectories outside of this sequence.

%C Even and odd terms alternate. No two squares can occur in succession because A106737 obtains even values for all squares > 1 and A000005 is odd for all squares. More directly this is seen from the fact that the rightmost 1-bit in the binary expansion of any square is always alone.

%C (End)

%H Yang Haoran, <a href="/A153013/b153013.txt">Table of n, a(n) for n = 0..23</a>

%F From _Antti Karttunen_, Oct 15 2016: (Start)

%F a(0) = 0; for n >= 1, a(n) = A005940(1+a(n-1)).

%F A008966(a(n+1)) = A085357(a(n)). [See the comment.]

%F A181819(a(1+n)) = A246029(a(n)).

%F A000005(a(n+1)) = A106737(a(n)).

%F (End)

%e 101 is interpreted as 3^1 * 2^1 = 6. 1110011 is interpreted as 5^3 * 2^2 = 500.

%t NestList[Times @@ Flatten@ MapIndexed[Prime[#2]^#1 &, #] &@ Flatten@ MapIndexed[If[Total@ #1 == 0, ConstantArray[0, Boole[First@ #2 == 1] + Length@ #1 - 1], Length@ #1] &, Reverse@ Split@ IntegerDigits[#, 2]] &, 0, 21] (* _Michael De Vlieger_, Oct 17 2016 *)

%o (PARI) step(n)=my(t=1,v); forprime(p=2,, v=valuation(n+1,2); t*=p^v; n>>=v+1; if(!n, return(t)))

%o t=0; concat(0,vector(20,n, t=step(t))) \\ _Charles R Greathouse IV_, Sep 01 2015

%o (Scheme, with memoization-macro definec)

%o (definec (A153013 n) (if (zero? n) n (A005940 (+ 1 (A153013 (- n 1))))))

%o ;; _Antti Karttunen_, Oct 15 2016

%Y Cf. A000005, A003714, A005940, A008966, A085357, A106737, A156552, A181819, A246029.

%K nonn

%O 0,3

%A Mark Zegarelli (mtzmtz(AT)gmail.com), Dec 16 2008

%E a(20)-a(22) from _Yang Haoran_, Aug 31 2015