🔢 mathematics

Dual $p$ -adic Diophantine approximation on manifolds

This paper resolves the homogeneous $p$ -adic convergence case of the Generalised Baker-Schmidt Problem for hypersurfaces of dimension at least three and other specific manifolds, while also providing slightly weaker results for the inhomogeneous setting, all without restricting approximation functions to be monotonic.

Original authors: Mumtaz Hussain, Johannes Schleischitz, Benjamin Ward

Published 2026-05-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mumtaz Hussain, Johannes Schleischitz, Benjamin Ward

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 you are trying to hit a moving target with a dart, but the target isn't just a single point; it's a complex, curved surface floating in a strange, digital universe called p-adic space. This paper is about figuring out exactly how "easy" or "hard" it is to hit that surface with a specific kind of mathematical dart.

Here is the breakdown of the paper's story, using simple analogies.

1. The Game: Throwing Darts at a Curved Surface

In the world of mathematics, there is a classic game called Diophantine Approximation.

The Target: A curved surface (called a manifold) inside a high-dimensional space. Think of this like a twisted ribbon floating in a room.
The Darts: These are simple whole numbers (integers).
The Goal: You want to see if you can get your "dart" (a combination of integers) to land incredibly close to a specific point on that curved ribbon.
The Rule: The closer you get, the better. But there's a catch: the "closer" you need to get gets harder and harder as you throw more darts.

The paper focuses on a specific version of this game played in p-adic space.

Real Space vs. p-adic Space: In our normal world (Real space), distance works like a ruler. In p-adic space, distance works like a "digital zoom" based on a specific prime number (like 2, 3, or 5). Two numbers are "close" if their difference is divisible by a high power of that prime. It's a very different, somewhat counter-intuitive way of measuring distance, but it's crucial for number theory.

2. The Big Question: How "Big" is the Set of Hits?

Mathematicians have known for a long time that if you throw enough darts, you will eventually hit the target. But the big question is: How "thick" is the set of points you can hit?

The "Zero" vs. "Full" Debate:
- If the rule for how close you need to be is very strict (you need to be extremely close), you might only hit a few points. In math terms, the "size" (measure) of these points is zero. It's like trying to hit a single hair on a giant balloon; you might hit it, but the area you cover is negligible.
- If the rule is loose (you just need to be somewhat close), you will hit almost everywhere. The "size" is full. It's like throwing darts blindfolded; you'll hit the balloon almost every time.

This is known as the Generalised Baker-Schmidt Problem. It's a famous puzzle asking: "At what exact point does the set of hits go from being 'nothing' to being 'everything'?"

3. What This Paper Solves

Previous researchers had solved parts of this puzzle for the "loose" rules (the divergence case) and for the "real world" version of the game. However, the strict case (the convergence case) in the p-adic world was a major open mystery.

The Authors' Achievement:
Mumtaz Hussain, Johannes Schleischitz, and Benjamin Ward have cracked the code for the strict case in p-adic space, but with some specific conditions:

The Surface Must Be Curved: They proved this for surfaces that are "curved enough" (specifically, hypersurfaces with a dimension of at least 3). If the surface is too flat, the math gets messy.
The Surface Must Be Smooth: In the p-adic world, the surface needs to be "analytic," which is a very strong type of smoothness (like a perfect, unbroken curve that can be described by a specific type of infinite formula).
The Result: They proved that if the "dart-throwing rule" is strict enough (mathematically, if a certain sum of numbers converges), then the set of points you can hit on the surface has zero size. In other words, you are almost guaranteed to miss the surface if the rules are that strict.

4. The "Magic" of the Proof

How did they prove this? They used a clever strategy involving magnifying glasses and covering.

The Problem: The surface is infinite and complex. You can't check every single point.
The Solution: They broke the surface down into tiny, manageable chunks (small balls).
The "Main Lemma": They proved a key rule (Lemma 2.3) that says: "If you are trying to hit a specific small chunk of the surface with a strict rule, the area you cover is so tiny that it disappears when you add up all the tiny pieces."
The Analogy: Imagine trying to cover a giant, curved wall with tiny stickers. If the rule for where you can put a sticker is very strict, the authors proved that even if you put stickers on every possible spot allowed, the total area of the stickers would still be effectively zero. The wall remains mostly uncovered.

5. Why Does This Matter?

The paper doesn't claim to cure diseases or build new computers. Its value is purely in mathematical truth.

It completes a major chapter in the story of Diophantine Approximation.
It connects the "real world" math (which we use for physics) with the "p-adic world" math (which is essential for cryptography and understanding the deep structure of numbers).
It shows that even in this strange, digital-like p-adic universe, the rules of geometry and approximation follow a predictable pattern: If the target is hard enough, you simply can't hit it often enough to make a "thick" set.

Summary

Think of this paper as a rigorous proof that if you make the rules of a game too hard, the winners become so rare that they effectively don't exist. The authors figured out exactly how to calculate that "rarity" for a specific, complex type of game played in a strange, number-theoretic universe. They showed that for curved surfaces in this universe, hitting the target under strict rules is a mathematical impossibility in terms of "size."

Technical Summary: Dual $p$ -adic Diophantine Approximation on Manifolds

Problem Statement
This paper addresses the Generalised Baker-Schmidt Problem (GBSP) within the context of dual Diophantine approximation on manifolds over the $p$ -adic numbers, $\mathbb{Q}_p$ . The GBSP seeks to determine the Hausdorff measure of the set of $\psi$ -approximable points on a nondegenerate manifold. While the divergence part of this problem (establishing that the measure is infinite when a certain series diverges) has been resolved for both real and $p$ -adic settings (by Beresnevich-Dickinson-Velani, Badziahin-Beresnevich-Velani, and Datta-Ghosh), the corresponding convergence part (establishing that the measure is zero when the series converges) remains a challenging open problem, particularly in the $p$ -adic setting.

The authors focus on the homogeneous case ( $\theta = 0$ ) and the inhomogeneous case ( $\theta \neq 0$ ) for dual approximation. Specifically, they investigate the set $D_n^{\theta}(\Psi)$ of points $x \in \mathbb{Q}_p^n$ satisfying $|a_1 x_1 + \dots + a_n x_n + a_0 + \theta|_p < \Psi(a)$ for infinitely many integer vectors $a = (a_0, \dots, a_n)$ . The goal is to prove that for specific classes of manifolds, the $f$ -dimensional Hausdorff measure of the intersection of this set with the manifold is zero, provided the sum of the approximation function terms converges.

Methodology
The paper employs a combination of $p$ -adic analysis, geometric measure theory, and the mass transference principle framework adapted to ultrametric spaces.

Analytic Manifolds and Strong Differentiability: Unlike the real setting where $C^2$ regularity often suffices, the authors require manifolds to be defined by analytic maps. This is necessitated by the $p$ -adic Inverse Function Theorem, which relies on strong differentiability (a stronger condition than ordinary differentiability in $p$ -adic spaces) to ensure local isomorphism properties.
Reduction to Local Estimates: The manifold $M$ is parameterized by an analytic map $g: U \subset \mathbb{Q}_p^d \to \mathbb{Q}_p^{n-d}$ . The problem is reduced to estimating the Hausdorff measure of the set $S(a) = \{x \in U : |h_a(x)|_p < \Psi(a)\}$ , where $h_a$ is a linear combination of the coordinates and the manifold function $g$ .
Key Lemma (Lemma 2.3): The core technical contribution is the proof of a covering lemma (Lemma 2.3) which bounds the Hausdorff content of $S(a)$ $S (a)$ . The proof involves a case-by-case analysis based on the gradient $\nabla h_a(x)$ $\nabla h_{a} (x)$ :
- Case (a): $a_2 = 0$ (linear part only).
- Case (b): The gradient is dominated by the linear term.
- Case (c): The gradient vanishes at a point (critical point).
- Case (d): The gradient is non-vanishing and bounded away from zero.
  In each case, the authors utilize the $p$ -adic Mean Value Theorem and the non-vanishing determinant of the Hessian matrix (under Condition II $_p$ ) to show that $S(a)$ can be covered by a specific number of balls of a certain radius.
Dimension Function Conditions: The proof relies on specific conditions for the dimension function $f(r)$ , denoted as (I $_p$ ) and (II $_p$ ). Condition (I $_p$ ) imposes a growth constraint ( $f(xy) \ll x^s f(y)$ ) for some $s < 2(d-1)$ . Condition (II $_p$ ) requires that the Hessian of the manifold parameterization is non-singular almost everywhere with respect to the Hausdorff measure $H_f$ .
Summation and Convergence: Using the Hausdorff-Cantelli Lemma, the authors sum the measures of the covering sets over the integer vectors. They carefully handle the summation over different subsets of integer vectors ( $Z^{(1)}$ and $Z^{(2)}$ ) and utilize the condition $\Psi(a) < \|a\|^{-1}$ to ensure the convergence of the series.

Key Contributions and Results
The primary result is Theorem 1.4, which establishes the convergence part of the GBSP for dual $p$ -adic approximation on manifolds under specific regularity and curvature conditions.

Theorem 1.4(i): For the set of integer vectors where $p$ divides at most one component (set $Z^{(1)}$ ), the Hausdorff measure $H_f(D_n^{\theta}(\Psi, Z^{(1)}) \cap M)$ is zero if $\sum \Psi(a)^{-(d-1)} f(\Psi(a)) < \infty$ . This holds for any dimension $d$ and satisfies the prediction of the GBSP.
Theorem 1.4(ii): For the set of integer vectors where $p$ does not divide all components (set $Z^{(2)}$ ), the measure is zero under the same convergence condition, provided $|\theta|_p \neq 1$ . The authors note this restriction on $\theta$ is an artifact of their method.
Theorem 1.4(iii): For the full set of non-zero integer vectors in the homogeneous case ( $\theta=0$ ), the measure is zero if the series converges and $\Psi$ satisfies a scaling condition ( $p\Psi(a) \leq \Psi(p^{-1}a)$ ).

Corollaries and Specific Cases:

Theorem 1.8: Combines the convergence result with the divergence result of Datta and Ghosh (Theorem 1.5) to provide a zero-full dichotomy for the $s$ -dimensional Hausdorff measure ( $d-1 < s < 2(d-1)$ ) on analytic nondegenerate manifolds.
Corollary 1.10: Explicitly calculates the Hausdorff dimension for power-law approximation functions $\Psi(a) = \|a\|^{-1-\tau}$ with $\tau > n$ . The dimension is given by $d - 1 + \frac{n+1}{1+\tau}$ .
Dimension $d=2$ : A notable feature of this work is that it extends results to hypersurfaces of dimension $d=2$ (ambient dimension 3), whereas previous real-setting results often required $d \geq 3$ . This is achieved by assuming strict inequality $\tau > n$ , which improves the parameter range for $s$ .

Significance and Claims
The paper claims to resolve the convergence counterpart of the Generalised Baker-Schmidt Problem for dual $p$ -adic approximation on a significant class of manifolds (hypersurfaces of dimension $d \geq 3$ and certain other classes satisfying Hessian conditions).

Extension of Theory: It extends the known divergence results (Datta-Ghosh) to a complete zero-full dichotomy for homogeneous dual approximation on sufficiently curved hypersurfaces in $p$ -adic space.
Regularity Requirements: The authors highlight that the requirement for analyticity (rather than just $C^2$ ) is a technical necessity in the $p$ -adic setting due to the nature of the Inverse Function Theorem in ultrametric spaces.
Limitations: The authors modestly note that the inhomogeneous result (Theorem 1.4(ii)) requires the condition $|\theta|_p \neq 1$ , for which they have no satisfactory explanation. Furthermore, the results for non-hypersurfaces (codimension $>1$ ) rely on Condition (II $_p$ ), which imposes rigid restrictions on the dimension pairs $(n, d)$ , though they suggest these might be simplifiable in the analytic case.
Scope: The results are presented for multivariable approximation functions $\Psi$ , though the corollaries often specialize to univariable forms. The paper does not claim to solve the GBSP for all manifolds or all approximation functions, but rather for those satisfying the specific analytic and curvature conditions outlined.