Nancy-Metz number theory seminar

One of the central topics in number theory is the study of $L$-functions and the distribution of their zeros. For example, the celebrated Prime Number Theorem is equivalent to the fact that the Riemann zeta function $\zeta(s)$ does not vanish on the line $\text{Re}(s)=1$. In this talk, we will focus on quadratic Dirichlet $L$-functions: in particular, the real zeros of the derivative of quadratic Dirichlet $L$-functions $L^\prime (s, \chi_d)$, where $d$ ranges over fundamental discriminants. Baker and Montgomery conjectured that there are $\asymp \log \log |d|$ real zeros of $L^\prime(s, \chi_d)$ in the interval $[1/2, 1]$ for almost all fundamental discriminants $d$. We will highlight some recent exciting progress that comes close to proving this conjecture and then outline the proof, which is based on ideas coming from analytic and probabilistic number theory. This is based on recent joint work with Youness Lamzouri.

Pour modéliser les premiers, nous considérerons au cours de l’exposé de telles questions de nature statistique concernant la suite des entiers sans facteur carré (SFC), parmi d’autres suites “criblées”. J’espère pouvoir motiver et expliquer notre résultat principal inconditionnel qui énonce que le nombre de SFC dans les intervalles courts uniformément aléatoires suit en effet une loi Gaussienne, ce faisant résolvant plusieurs problèmes de R.R. Hall.

Ceci est un travail en commun avec O. Gorodetsky et B. Rodgers.

We present a new elementary algorithm for computing $M(x) = \sum_{n \leq x} \mu(n),$ where $\mu(n)$ is the Möbius function. Our algorithm takes
\[\begin{aligned}
\mathrm{time} \ \ O_\epsilon\left(x^{\frac{3}{5}} (\log x)^{\frac{3}{5}+\epsilon} \right)
\ \ \mathrm{and}\ \ \mathrm{space} \ \ O\left(x^{\frac{3}{10}} (\log x)^{\frac{13}{10}}
\right)\end{aligned},\] which improves on existing combinatorial algorithms. While there is an analytic algorithm due to Lagarias-Odlyzko with computations based on the integrals of $\zeta(s)$ that only takes time $O(x^{1/2 + \epsilon})$, our algorithm has the advantage of being easier to implement. The new approach roughly amounts to analyzing the difference between a model that we obtain via Diophantine approximation and reality, and showing that it has a simple description in terms of congruence classes and segments. This simple description allows us to compute the difference quickly by means of a table lookup. This talk is based on joint work with Harald Andrés Helfgott.

Given an arithmetic function $\theta$, we consider the set
$$ \mathcal{B}_\theta = \Bigl\{n\ge 1: p|n \Rightarrow p\le \theta\Bigl(\prod_{q<p \atop q^\alpha || n} q^\alpha \Bigr) \Bigr\},$$
where $p$ and $q$ denote primes. Depending on the choice of $\theta$, the possible sets $\mathcal{B}_\theta$ include the set of prime powers, almost primes, friable numbers, dense numbers, and practical numbers.
We will discuss (1) asymptotic results for the counting function of $\mathcal{B}_\theta$, (2) a generalization of the Siegel-Walfisz theorem, and (3) the normal order of the number of prime factors of integers in $\mathcal{B}_\theta$.

Considering partial character sums as defining a family of random processes (by choosing the characters randomly from some set), it becomes natural to ask questions about the limiting distribution. I’ll describe this in some contexts and give examples of what we can find in the support. This is largely work of Ayesha Hussain, but also some joint work.

Soit P un polynôme à coefficients entiers, unitaire, irréductible, de degré 4 et de groupe de Galois diédral ou cyclique.
Il existe c = c(P) > 0, tel que P(n) ait un facteur premier supérieur à n^1+c pour une proportion positive d’entiers n.

Il s’agit d’un travail  avec James Maynard.

Répétition du séminaire Bourbaki du vendredi 1er avril.

Chebyshev famously observed empirically that more often than not, there are more primes of the form 3 mod 4 up to x than primes of the form 1 mod 4. This was confirmed theoretically much later by Rubinstein and Sarnak in a logarithmic density sense. Our understanding of this is conditional on the generalized Riemann Hypothesis as well as Linear Independence of the zeros of L-functions.

We investigate similar questions for sums of two squares in arithmetic progressions. We find a significantly stronger bias than in primes, which happens for almost all integers in a natural density sense. Because the bias is more pronounced, we do not need to assume Linear Independence of zeros, only a Chowla-type Conjecture on non-vanishing of L-functions at 1/2.

We’ll aim to be self-contained and define all the notions mentioned above during the talk. We shall review the origin of the bias in the case of primes and the work of Rubinstein and Sarnak. We’ll explain the main ideas behind the proof of the bias in the sums-of-squares setting.

Récemment, Lemke Oliver et Soundararajan ont observé d’importants biais dans la répartition de couples de nombres premiers consécutifs dans les progressions arithmétiques. Ils ont proposé un modèle heuristique basé sur la conjecture de Hardy–Littlewood qui explique très bien ces observations.
Nous discuterons la question analogue pour les nombres qui s’écrivent comme une somme de deux carrés d’entiers. Un biais semblable apparaît dans les données et nous développons un modèle heuristique similaire pour l’expliquer.
Travail joint avec Chantal David, Jungbae Nam et Jeremy Schlitt.