login

Year-end appeal: Please make a donation to the OEIS Foundation to support ongoing development and maintenance of the OEIS. We are now in our 61st year, we have over 378,000 sequences, and we’ve reached 11,000 citations (which often say “discovered thanks to the OEIS”).

k-digit integers equal to the sum of the k-th powers of the tens' complements of their digits.
3

%I #41 Jul 10 2018 10:15:50

%S 5,378,91882,3762938,46478818,564426414

%N k-digit integers equal to the sum of the k-th powers of the tens' complements of their digits.

%C The terms of the sequence could be called "Shy n n-digit numbers" as suggested by Geoffrey Campbell, Long Term Visitor (Visiting Fellow), Mathematical Sciences Institute, Australian National University, cf Links.

%C In base 10, x is a "Shy k n-digit number" if it is an n-digit (d_i) number such that x = Sum_{i=1..n}{(10-d_i)^k}. For instance, 2240 is a "Shy 3 4-digit number": (10 - 2)^3 + (10 - 2)^3 + (10 - 4)^3 + (10 - 0)^3 = 512 + 512 + 216 + 1000 = 2240. Again, 2149042 is a "Shy 6 7-digit number": (10 - 2)^6 + (10 - 1)^6 + (10 - 4)^6 + (10 - 9)^6 + (10 - 0)^6 + (10 - 4)^6 + (10 - 2)^6 = 262144 + 531441 + 46656 + 1 + 1000000 + 46656 + 262144 = 2149042.

%C It is not known if the sequence is finite. At least there are no other terms up to 18-digit numbers (as tested by Marco Cecchi at LinkedIn link).

%C If there are further terms, they are greater than 10^33. - _Giovanni Resta_, Aug 20 2015

%C Subsequence of A052382. Sequence is finite and complete as verified by exhaustive search since all terms have 60 or fewer digits. Since all terms are zeroless, they are less than k*9^k which would be less than 10^(k-1) (i.e., have fewer than k digits) if k > 60. - _Chai Wah Wu_, Apr 07 2018

%H Geoffrey Campbell, <a href="https://www.linkedin.com/grp/post/4510047-6038963758880010243"> Related to Narcissistic numbers: the Shy numbers</a>, Number Theory group on LinkedIn.com

%H Marco Cecchi, <a href="/A261433/a261433.txt">Python program based on partitions</a>.

%e (10 - 5)^1 = 5,

%e (10 - 3)^3 + (10 - 7)^3 + (10 - 8)^3 = 343 + 27 + 8 = 378,

%e (10 - 9)^5 + (10 - 1)^5 + (10 - 8)^5 + (10 - 8)^5 + (10 - 2)^5 = 1 + 59049 + 32 + 32 + 32768 = 91882, etc.

%p with(numtheory): P:=proc(q) local a,b,c,k,n;

%p for n from 1 to q do a:=ilog10(n)+1; b:=0; c:=n;

%p for k from 1 to a do b:=b+(10-(c mod 10))^a; c:=trunc(c/10); od;

%p if b=n then print(n); fi; od; end: P(10^9);

%t Select[Range[10^5], # == Total[(10 - IntegerDigits@ #)^ IntegerLength[#]] &] (* _Giovanni Resta_, Aug 20 2015 *)

%o (PARI) isok(n) = (d = digits(n)) && (sum(k=1, #d, (10-d[k])^#d) == n); \\ _Michel Marcus_, Aug 24 2015

%o (Python)

%o from itertools import combinations_with_replacement

%o A261433_list = []

%o for k in range(1,10):

%o a, k10 = tuple([i**k for i in range(10,0,-1)]), 10**k

%o for b in combinations_with_replacement(range(1,10),k):

%o x = sum(list(map(lambda y:a[y],b)))

%o if x < k10 and tuple(int(d) for d in sorted(str(x))) == b:

%o A261433_list.append(x)

%o A261433_list = sorted(A261433_list) # _Chai Wah Wu_, Aug 25 2015, updated Apr 06, 2018

%Y Cf. A005188, A052382.

%K nonn,base,fini,full

%O 1,1

%A _Paolo P. Lava_, Aug 20 2015

%E a(4)-a(6) found by Aleksander Zujev