Variants on the $abc$-Conjecture using Alternative Quality Metrics

This paper introduces new variants of the $abc$-conjecture based on doubly geometric mean quality metrics, establishing asymptotic results and sharp phase transitions for high-quality triples while developing sub-linear algorithms to efficiently identify them.

The Gist

Imagine you are a detective trying to solve a mystery about numbers. In the world of mathematics, there is a famous, unsolved puzzle called the abc-conjecture. It's like a rulebook that tries to predict how "special" certain groups of three numbers are when they add up together (where $a + b = c$).

For decades, mathematicians have tried to measure how "special" these groups are using a specific ruler called the Standard Quality. Think of this ruler as a scale that weighs the size of the numbers against the complexity of their prime building blocks. The rulebook says that while you can find some very special groups, they shouldn't get too special, too often. But proving this has been incredibly difficult, like trying to catch a ghost.

The New Idea: A Different Ruler
In this paper, the author, Akilan Sankaran, decides to try a different approach. Instead of using the old ruler, he invents a new set of rulers called DGM Quality Metrics.

Here is the core difference:

• The Old Ruler cares a lot about the size of the prime numbers involved.
• The New Ruler is obsessed with the count of prime numbers. It has a "penalty system." If a group of numbers is built from a huge pile of different prime ingredients, the new ruler says, "You are not very special," even if the numbers themselves are huge. It prefers groups built from just a few, simple ingredients.

The "Double-Geometric" Twist
To make this new ruler work, the author uses a mathematical tool called the Doubly Geometric Mean.

• Imagine you have a bag of numbers.
• The Arithmetic Mean is like adding them all up and dividing by the count (the average).
• The Geometric Mean is like multiplying them and taking a root (a way to find a "typical" size that handles big numbers better).
• The Doubly Geometric Mean is like taking the geometric mean of the logarithms of the numbers. It's a super-sensitive filter that shrinks the influence of large numbers even more, making the number of ingredients (the count of primes) the most important factor.

What Did They Find?
Using this new ruler, the author discovered some surprising things:

1. The Limit Doesn't Exist (It Goes to Infinity): Unlike the old rulebook which suggests there's a limit to how special a group can be, the new ruler shows that if you look at specific families of numbers (like those related to Mersenne primes, which are special numbers of the form $2^n - 1$), the "quality" score can grow forever. It's like finding a ladder that goes up to the sky; the higher you climb, the more special the numbers get.
2. Phase Transitions (The Tipping Point): The author created a tunable version of this ruler with knobs labeled $\alpha$ and $\beta$. By turning these knobs, they found "tipping points."
   • If you turn the knob one way, the scores go to infinity (unbounded).
   • If you turn it another way, the scores crash down to zero.
   • There is a "Goldilocks" setting in the middle where the scores settle at a specific, finite number. This is like finding the perfect temperature where water turns into ice; the behavior of the numbers changes drastically depending on how you tune the metric.
3. Connecting the Dots: The author showed that this new ruler is actually mathematically linked to the old one. It's like having a new map that helps you navigate the same territory. By understanding how the new ruler behaves, we might get a better clue about how the old, difficult rulebook works.

The Speed Boost: Finding the Winners Faster
One of the biggest practical achievements in the paper is the creation of new algorithms (step-by-step computer instructions).

• The Old Way: To find the best number groups, computers used to check almost every possibility one by one. This is like searching for a needle in a haystack by looking at every single piece of straw. It's slow and gets impossible as the haystack grows.
• The New Way: The author's new algorithms act like a metal detector that only beeps for the needle. By focusing on specific families of numbers (like powers of 2 and 3, or Mersenne primes), the computer skips the haystack entirely.
• The Result: The new method is incredibly fast. It can find number groups with record-breaking quality scores that have millions of digits, whereas the old method would take longer than the age of the universe to find them.

In Summary
This paper doesn't solve the original abc-conjecture. Instead, it builds a new, more flexible laboratory to study the problem. By changing the rules of how we measure "specialness" to focus on the count of prime factors rather than just their size, the author:

1. Proved that under these new rules, "special" numbers can get infinitely special.
2. Mapped out exactly where the behavior of these numbers changes (the phase transitions).
3. Built super-fast tools to find these special numbers, allowing us to explore mathematical territory that was previously unreachable.

It's a new lens through which to view an old mystery, offering fresh insights and powerful tools to keep digging.

Technical Summary

Technical Summary: Variants on the abc-Conjecture using Alternative Quality Metrics

Problem Statement
The abc-conjecture, formulated by Masser and Oesterl´e, posits an asymptotic bound on the "quality" $q_s(a, b, c) = \ln(c) / \ln(\text{rad}(abc))$ of coprime triples $(a, b, c)$ satisfying $a+b=c$. While the conjecture implies profound results in number theory (including Fermat's Last Theorem for $n \ge 6$), it remains unproven. A primary difficulty in analyzing the conjecture is that determining high-quality triples under the standard metric is computationally intractable for large $c$, and the standard metric prioritizes triples with small radicals, which can be achieved by triples with many prime factors if exponents are sufficiently large. This paper investigates whether alternative quality metrics, specifically those that penalize the number of distinct prime factors ($\omega$) more aggressively, can yield tractable variants of the conjecture with analogous asymptotic behaviors.

Methodology
The authors introduce a new class of quality metrics based on the Doubly Geometric Mean (DGM) of the prime factors of $abc$.

1. Definition of DGM Quality: The standard quality is viewed as a comparison of $\ln(c)$ to the geometric mean of the primes dividing $abc$. The authors define the DGM quality, $q_{DGM}$, by replacing the geometric mean with the doubly geometric mean:
   $$q_{DGM}(a, b, c) = \frac{\ln(c)}{\exp\left(\frac{1}{\omega} \sum_{i=1}^\omega \ln \ln p_i\right)} = \frac{\ln(c)}{\omega \cdot \sqrt[\omega]{\prod_{i=1}^\omega \ln p_i}}$$
   This metric explicitly penalizes the number of distinct prime factors $\omega$ via a linear factor in the denominator and logarithmic weighting of the primes, effectively prioritizing triples with few prime factors regardless of their magnitude.

2. Parametrized Class: The authors generalize this into a DGM Quality Class $q_C(a, b, c; \alpha, \beta)$, introducing parameters $\alpha$ and $\beta$ to tune the influence of $\omega$ and the growth rate of the denominator, respectively.

3. Analytical Approach: The paper employs:

   • Asymptotic Analysis: Constructing specific families of triples (e.g., Mersenne, Twin Prime, Fixed Prime Sequences) to determine the $\limsup$ of the quality metrics.
   • Conditional Proofs: Leveraging conjectures (Twin Prime, Lenstra-Pomerance-Wagstaff) and theorems (Chen's Theorem, Mihailescu's Theorem) to establish bounds.
   • Connection to Elliptic Curves: Mapping abc-triples to Frey curves to relate the new metrics to the Szpiro ratio and the standard abc-conjecture.
   • Algorithmic Implementation: Developing constructive enumeration algorithms to find high-quality triples with sub-linear runtime compared to brute-force approaches.

Key Contributions and Results

• Unboundedness of DGM Quality: Unlike the standard quality (conjectured to be bounded by 1), the authors prove that $q_{DGM}$ is unbounded.

  • Theorem 3.9: $\limsup_{c \to \infty} \max q_{DGM} = \infty$.
  • This is established conditionally on the infinitude of Mersenne primes (Theorem 3.8), Fermat primes, or Twin primes (Theorem 3.10).
  • Theorem 3.12: An unconditional proof of the unboundedness is provided using Chen's Theorem, constructing a family of triples $(2, p, p+2)$ where $p+2$ is either prime or a product of two primes ($P_2$).

• Phase Transitions in the DGM Class: The paper identifies critical thresholds for the parameter $\beta$ in the class $q_C$.

  • Theorem 4.4: For families of triples with a fixed number of primes $\omega$ and $k$ primes growing with $c$, a phase transition occurs at $\beta_0 = \omega/k$.
    • If $\beta < \omega/k$, the quality diverges to $\infty$.
    • If $\beta = \omega/k$, the quality converges to a finite non-zero constant.
    • If $\beta > \omega/k$, the quality converges to 0.
  • This establishes a "saturation" condition where increasing $\beta$ prunes specific families of triples from the set of high-quality candidates.

• Relationship to Standard Quality:

  • Theorem 4.13 & 4.15: The standard quality $q_s$ is shown to be related to the DGM quality via a "packing efficiency" factor $\eta$ (the ratio of the geometric mean to the arithmetic mean of the logarithms of the primes). Specifically, $q_s = \eta \cdot q_{DGM}$.
  • The authors show that the abc-conjecture is equivalent to the assertion that $q_{DGM}$ is bounded by $1/\eta$ (up to a constant), suggesting that the divergence of $q_{DGM}$ is controlled by the decay of the packing efficiency $\eta$ as the number of primes increases.

• Algorithmic Efficiency:

  • The paper presents four algorithms for finding high-quality triples.
  • Algorithms 3 & 4 (Mersenne and Cyclotomic): These utilize precomputed factorizations of numbers of the form $b^n - 1$. They achieve $O(d)$ complexity (where $d$ is the number of digits of $c$), a significant improvement over the $O(10^d)$ complexity of brute-force methods.
  • Using the Cyclotomic Algorithm (Algorithm 4) with base $b=7$, the authors identified a triple with a DGM quality of $\approx 14,060$, far exceeding the highest Mersenne quality ($\approx 4,544$), despite having a smaller $c$.

Significance and Claims
The paper claims that the DGM metrics offer a "robust variation" of the abc-conjecture that is analytically more tractable.

• Tractability: By shifting the focus to metrics that penalize $\omega$ heavily, the authors demonstrate that the analogue of the abc-conjecture (unboundedness of quality) can be proven unconditionally using Chen's Theorem, whereas the standard conjecture remains open.
• Phase Space Analysis: The introduction of the $(\alpha, \beta)$ parameter space allows for a granular study of how different families of triples behave asymptotically. The identification of phase transitions suggests that the standard conjecture might be viewed as a specific point within a continuum of quality metrics.
• Computational Advantage: The development of sub-linear algorithms allows for the exploration of much larger triples than previously possible under the standard metric, providing empirical data on the distribution of high-quality triples under the new metrics.
• Theoretical Bridge: The work establishes a direct link between the standard abc-conjecture and the DGM metrics via the packing efficiency of primes, suggesting that understanding the asymptotic behavior of the DGM class could provide insights or "damping" mechanisms relevant to proving the original conjecture.

The authors conclude that while the DGM analogue is unbounded (and thus distinct from the bounded standard conjecture), the framework of modified quality metrics and their phase transitions offers a promising avenue for independent analytical questions and potentially deeper understanding of the prime structure underlying the abc-conjecture.