Abstract:We report the finding of the new upper bound on the lowest positive integer $x$ for which the Mertens conjecture \begin{equation*} \left| \sum_{1 \leq n \leq x} \mu(n) \right| < \sqrt{x} \end{equation*} fails to hold: $x < \exp(1.017 \times 10^{29})$, an improvement over previously known $\exp(1.59 \times 10^{40})$ due to Kotnik and te Riele [7].

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the upper bound estimation of the smallest counterexample regarding Mertens' conjecture. Specifically, Mertens' conjecture was proposed by Mertens in 1897, and its content is: \[ M(x) := \left| \sum_{1 \leq n \leq x} \mu(n) \right| < \sqrt{x} \] Here, $\mu(n)$ is the Möbius function. This conjecture lasted for nearly a century until it was disproved by Odlyzko and te Riele in 1985 through the application of the LLL algorithm. Although Mertens' conjecture has been proven false, two natural follow - up questions still remain: 1. What is the correct asymptotic growth rate of $M(x)$? 2. What is the smallest $x$ such that $|M(x)| \geq \sqrt{x}$? This paper mainly focuses on the second question and attempts to find a smaller upper bound. The previously best - known upper bound was given by Kotnik and te Riele in 2006: \[ x < \exp(1.59\times 10^{40}) \] And this paper, through the improved lattice - basis reduction algorithm (especially the BKZ algorithm), improves this upper bound to: \[ x < \exp(1.017\times 10^{29}) \] In addition, the author also found some suggestive data points that are close to the size of the smallest counterexample speculated by Kotnik and van de Lune: \[ x \approx \exp(5.15\times 10^{23}) \] Overall, the main contribution of this paper lies in using modern lattice - basis reduction techniques to significantly reduce the upper bound estimation of the smallest counterexample of Mertens' conjecture.

A new upper bound on the smallest counterexample to the Mertens conjecture

New explicit bounds for Mertens function and the reciprocal of the Riemann zeta-function

Higher Mertens constants for almost primes II

Multiple Mertens evaluations

Improving bounds on prime counting functions by partial verification of the Riemann hypothesis

A counterexample to the Conjecture of Ankeny, Artin and Chowla

New bounds and progress towards a conjecture on the summatory function of $(-2)^{Ω(n)}$

A New Strategy for Riemann Hypothesis

On proving an Inequality of Ramanujan using Explicit Order Estimates for the Mertens Function

A New Theorem on the Prime-Counting Function

On the mean of the error terms in Mertens' theorems

Uniform bounds for Kloosterman sums of half-integral weight with applications

Improved Bounds for Szemerédi's Theorem

New Lower Bounds for the Schur-Siegel-Smyth Trace Problem

D{é}monstration d'une conjecture de Kruyswijk et Meijer sur le plus petit d{é}nominateur des nombres rationnels d'un intervalle

An Analytical Exploration of the Erdös-Moser Equation $ \sum_{i=1}^{m-1} i^k = m^k $ Using Approximation Methods

Sharper bounds for the error in the prime number theorem assuming the Riemann Hypothesis

On a conjecture of R. M. Murty and V. K. Murty II

From explicit estimates for the primes to explicit estimates for the Möbius function II

Mertens' prime product formula, dissected

On the Brun--Titchmarsh theorem