Abstract: I'll discuss joint work with Kaisa Matomaki on sign changes of Fourier coefficients of eigencuspforms. I'll motivate the problem and I will explain how and why our work depends on sieve-theoretic analogues of mollifiers. Mollifiers are typically used to understand the zero-distribution of L-functions.
Abstract: We discuss Tenenbaum's proof of an old conjecture of Erdos about the sum E(n):=∑i>j1/(di - dj), where d1 < d2<... are the divisors of an integer n. Erdos conjectured an upper bound of d(n) for this sum, where d(n) is the number of divisors of n; Tenenbaum was able to prove a strong form of this conjecture, with a bound of E(n)<< d(n)1-c, with c>5/83.
Given an integer k≥2 the function dk(n) equals the number of ways of writing n as a product of k positive integers. The variance of sums of the divisor function d2(n)=d(n) in short intervals was studied by Matti Jutila and later by Aleksandar Ivic. I will discuss work in progress on the variance of sums of dk(n) in short intervals for k=3. Additionally, I will indicate relationships between this variance and the 2k-th moment of the Riemann zeta-function that I am still trying to understand.
Contact me at: rudnick@post.tau.ac.il, Office : Schreiber 316, tel: 640-7806