← Latest papers
🔢 mathematics

Correlations of multiplicative functions with their partial sums

Assuming the Riemann hypothesis and the simplicity of the nontrivial zeros of the Riemann zeta function, this paper establishes asymptotic formulas for the logarithmic correlations between the Möbius and Liouville functions and their respective partial sums, revealing an anticorrelation that suggests effective upper bounds on the reciprocal of the derivative of the zeta function at its zeros.

Original authors: Gordon Chavez

Published 2026-05-15
📖 4 min read🧠 Deep dive

Original authors: Gordon Chavez

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the number line as a vast, infinite ocean. In this ocean, there are two very special, mysterious creatures: the Möbius function (μ\mu) and the Liouville function (λ\lambda).

These creatures are "multiplicative," meaning their behavior depends on the prime factors of the numbers they visit. But here's the twist: they are incredibly chaotic. They jump up and down between positive and negative values in a way that looks completely random, like a drunk sailor stumbling across a deck.

Now, imagine we have a second creature for each of them: a Summatory Function (MM for Möbius, LL for Liouville). Think of these as "memory keepers." Every time the chaotic creature jumps, the memory keeper adds that jump to a running total. So, M(n)M(n) is the total sum of all the Möbius jumps from 1 up to nn.

The Big Question: Do They Dance Together?

The paper asks a fascinating question: Does the chaotic creature (μ\mu or λ\lambda) have a secret relationship with its own memory keeper (MM or LL)?

Specifically, if we look at the chaotic creature at step nn, and compare it to the memory keeper's total just before that step (n1n-1), do they move in sync (correlation) or do they move in opposite directions (anticorrelation)?

Think of it like a dance:

  • Correlation: If the dancer steps left, the partner steps left.
  • Anticorrelation: If the dancer steps left, the partner steps right.

The Author's Discovery

The author, Gordon Chavez, uses a special mathematical "magnifying glass" (called a logarithmic average) to zoom in on this relationship over a huge range of numbers. He assumes two big, unproven but widely believed ideas in math:

  1. The Riemann Hypothesis: The "hidden rhythm" of prime numbers is perfectly regular.
  2. Simple Zeros Conjecture: Every note in that rhythm is unique (no two notes are exactly the same pitch).

Under these assumptions, the paper finds a surprising result: The dance is an anticorrelation.

  • For the Möbius function: When the chaotic creature jumps, the memory keeper tends to be moving in the opposite direction. The paper calculates a specific "negative value" for this relationship, suggesting they are constantly pushing against each other.
  • For the Liouville function: The same thing happens. The chaotic creature and its memory keeper are also anticorrelated, though the math is slightly different because the Liouville creature has a built-in "bias" (it tends to lean slightly negative on average).

The "Ghost" in the Machine

Why does this matter? The paper suggests that this anticorrelation is caused by the "ghosts" of the Riemann zeta function's zeros.

Imagine the zeros of the zeta function as invisible pillars holding up the structure of the number system. The paper shows that the strength of the anticorrelation is directly tied to how "tall" or "strong" these pillars are.

If the anticorrelation is real and strong (as the numbers suggest), it implies that these invisible pillars cannot be too weak. In mathematical terms, this gives us a new way to estimate the size of a very tricky number called 1/ζ(ρ)1/\zeta'(\rho).

The "Mertens" Warning

The paper ends with a cautionary tale. It mentions that in the past, mathematicians looked at similar numbers and thought they saw a pattern (that the memory keeper never got too big). They were wrong. A counterexample was eventually found, proving that even if a pattern looks perfect for millions of numbers, it can suddenly break.

So, while the paper presents strong evidence that these functions are anticorrelated and that this gives us new limits on the "ghost pillars," it reminds us that in the world of infinite numbers, we must always be careful not to trust our eyes too much.

Summary in a Nutshell

  • The Characters: Chaotic number functions and their running totals.
  • The Action: Comparing the current step to the previous total.
  • The Result: They move in opposite directions (anticorrelation).
  • The Implication: This movement reveals hidden constraints on the fundamental structure of prime numbers, specifically regarding the "strength" of the Riemann zeta function's zeros.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →