MathLabs

Waring's problem

Solved, 1909Arithmetic and number theory
Statement

For every integer exponent k≥1k \ge 1, there exists a finite number g(k)g(k) such that every positive integer n≥1n \ge 1 can be expressed as the sum of at most s≤g(k)s \le g(k) non-negative kk-th powers, n=x1k+x2k+⋯+xskn = x_1^k + x_2^k + \dots + x_s^k.

Originally posed by Edward Waring in 1770, the qualitative existence of g(k)g(k) for every kk (the Hilbert–Waring theorem) was proved by David Hilbert in 1909 using polynomial identities derived from 2525-fold integrals over the unit sphere. In the 1920s, G. H. Hardy and J. E. Littlewood developed the analytic circle method to give quantitative bounds and study G(k)G(k)—the number of kk-th powers needed for all sufficiently large integers—later refined by I. M. Vinogradov, Hua Luogeng, and Trevor Wooley.

Because small integers like 2k⌊(3/2)k⌋−12^k \lfloor (3/2)^k \rfloor - 1 can only be summed using 1k1^k and 2k2^k, J. A. Euler conjectured in 1772 that g(k)=2k+⌊(3/2)k⌋−2g(k) = 2^k + \lfloor (3/2)^k \rfloor - 2 for every kk, which is proved for all known values and holds unless {(3/2)k}>1−(3/4)k\{(3/2)^k\} > 1 - (3/4)^k. The more subtle asymptotic quantity G(k)G(k) (for all sufficiently large integers) is known exactly only for G(2)=4G(2) = 4 and G(4)=16G(4) = 16 (Davenport, 1939), while even 4≤G(3)≤74 \le G(3) \le 7 remains open.

References

  1. David Hilbert (1909). Beweis für die Darstellbarkeit der ganzen Zahlen durch eine feste Anzahl n-ter Potenzen (Waringsches Problem) · DOI:10.1007/BF01450405
  2. R. C. Vaughan, Trevor D. Wooley (2002). Waring's problem: a survey