The conditional-convergence conjecture for coefficients of L-functions

Let L(s)=b(n)/nsL(s)=\sum b(n)/n^s be an entire LL-function of degree kk, normalized so that its critical line is Re(s)=1/2\operatorname{Re}(s)=1/2. Let PLP_L denote a polynomial. More generally, suppose that L(s)L(s) is meromorphic and has only a pole at s=1s=1, of order rr.

Conditional-convergence conjecture. For an entire LL-function,

nXb(n)=O(X(k1)/2k+ϵ).\sum_{n\leq X}b(n)=O\left(X^{(k-1)/2k+\epsilon}\right).

More generally, for a meromorphic LL-function as above,

nXb(n)=XPL(logX)+O(X(k1)/2k+ϵ),\sum_{n\leq X}b(n)=XP_L(\log X)+O\left(X^{(k-1)/2k+\epsilon}\right),

where PLP_L is a polynomial of degree r1r-1.

This is proposed by analogy with the conjectured remainder size in the kk-divisor problem: after removing the polynomial main term caused by the pole, the coefficients should exhibit cancellation of order X(k1)/2k+ϵX^{(k-1)/2k+\epsilon}.

Sources & referencesView supporting material

Primary source

Michael O. Rubinstein, “The conditional convergence of the Dirichlet series of an L-function”, arXiv:0811.4241 (2009).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.