<p>We study higher uniformity properties of the von Mangoldt function <InlineEquation ID="IEq1"> <EquationSource Format="MATHML"><math> <mi mathvariant="normal">Λ</mi> </math></EquationSource> <EquationSource Format="TEX">$\Lambda $</EquationSource> </InlineEquation>, the Möbius function <InlineEquation ID="IEq2"> <EquationSource Format="MATHML"><math> <mi>μ</mi> </math></EquationSource> <EquationSource Format="TEX">$\mu $</EquationSource> </InlineEquation>, and the divisor functions <InlineEquation ID="IEq3"> <EquationSource Format="MATHML"><math> <msub> <mi>d</mi> <mi>k</mi> </msub> </math></EquationSource> <EquationSource Format="TEX">$d_{k}$</EquationSource> </InlineEquation> on short intervals <InlineEquation ID="IEq4"> <EquationSource Format="MATHML"><math> <mo stretchy="false">(</mo> <mi>x</mi> <mo>,</mo> <mi>x</mi> <mo>+</mo> <mi>H</mi> <mo stretchy="false">]</mo> </math></EquationSource> <EquationSource Format="TEX">$(x,x+H]$</EquationSource> </InlineEquation> for almost all <InlineEquation ID="IEq5"> <EquationSource Format="MATHML"><math> <mi>x</mi> <mo>∈</mo> <mo stretchy="false">[</mo> <mi>X</mi> <mo>,</mo> <mn>2</mn> <mi>X</mi> <mo stretchy="false">]</mo> </math></EquationSource> <EquationSource Format="TEX">$x \in [X, 2X]$</EquationSource> </InlineEquation>. Let <InlineEquation ID="IEq6"> <EquationSource Format="MATHML"><math> <msup> <mi mathvariant="normal">Λ</mi> <mi mathvariant="normal">♯</mi> </msup> </math></EquationSource> <EquationSource Format="TEX">$\Lambda ^{\sharp }$</EquationSource> </InlineEquation> and <InlineEquation ID="IEq7"> <EquationSource Format="MATHML"><math> <msubsup> <mi>d</mi> <mi>k</mi> <mi mathvariant="normal">♯</mi> </msubsup> </math></EquationSource> <EquationSource Format="TEX">$d_{k}^{\sharp }$</EquationSource> </InlineEquation> be suitable approximants of <InlineEquation ID="IEq8"> <EquationSource Format="MATHML"><math> <mi mathvariant="normal">Λ</mi> </math></EquationSource> <EquationSource Format="TEX">$\Lambda $</EquationSource> </InlineEquation> and <InlineEquation ID="IEq9"> <EquationSource Format="MATHML"><math> <msub> <mi>d</mi> <mi>k</mi> </msub> </math></EquationSource> <EquationSource Format="TEX">$d_{k}$</EquationSource> </InlineEquation>, <InlineEquation ID="IEq10"> <EquationSource Format="MATHML"><math> <mi>G</mi> <mo stretchy="false">/</mo> <mi mathvariant="normal">Γ</mi> </math></EquationSource> <EquationSource Format="TEX">$G/\Gamma $</EquationSource> </InlineEquation> a filtered nilmanifold, and <InlineEquation ID="IEq11"> <EquationSource Format="MATHML"><math> <mi>F</mi> <mo>:</mo> <mi>G</mi> <mo stretchy="false">/</mo> <mi mathvariant="normal">Γ</mi> <mo stretchy="false">→</mo> <mi mathvariant="double-struck">C</mi> </math></EquationSource> <EquationSource Format="TEX">$F \colon G/\Gamma \to \mathbb{C}$</EquationSource> </InlineEquation> a Lipschitz function. Then our results imply for instance that when <InlineEquation ID="IEq12"> <EquationSource Format="MATHML"><math> <msup> <mi>X</mi> <mrow> <mn>1</mn> <mo stretchy="false">/</mo> <mn>3</mn> <mo>+</mo> <mi>ε</mi> </mrow> </msup> <mo>≤</mo> <mi>H</mi> <mo>≤</mo> <mi>X</mi> </math></EquationSource> <EquationSource Format="TEX">$X^{1/3+\varepsilon } \leq H \leq X$</EquationSource> </InlineEquation> we have, for almost all <InlineEquation ID="IEq13"> <EquationSource Format="MATHML"><math> <mi>x</mi> <mo>∈</mo> <mo stretchy="false">[</mo> <mi>X</mi> <mo>,</mo> <mn>2</mn> <mi>X</mi> <mo stretchy="false">]</mo> </math></EquationSource> <EquationSource Format="TEX">$x \in [X, 2X]$</EquationSource> </InlineEquation>, <Equation ID="Equa"> <EquationSource Format="MATHML"><math> <munder> <mo movablelimits="false">sup</mo> <mrow> <mi>g</mi> <mo>∈</mo> <mo>Poly</mo> <mo stretchy="false">(</mo> <mi mathvariant="double-struck">Z</mi> <mo stretchy="false">→</mo> <mi>G</mi> <mo stretchy="false">)</mo> </mrow> </munder> <mrow> <mo>|</mo> <munder> <mo movablelimits="false">∑</mo> <mrow> <mi>x</mi> <mo>&lt;</mo> <mi>n</mi> <mo>≤</mo> <mi>x</mi> <mo>+</mo> <mi>H</mi> </mrow> </munder> <mo stretchy="false">(</mo> <mi mathvariant="normal">Λ</mi> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> <mo>−</mo> <msup> <mi mathvariant="normal">Λ</mi> <mi mathvariant="normal">♯</mi> </msup> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> <mo stretchy="false">)</mo> <mover accent="true"> <mi>F</mi> <mo>‾</mo> </mover> <mo stretchy="false">(</mo> <mi>g</mi> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> <mi mathvariant="normal">Γ</mi> <mo stretchy="false">)</mo> <mo>|</mo> </mrow> <mo>≪</mo> <mi>H</mi> <msup> <mo>log</mo> <mrow> <mo>−</mo> <mi>A</mi> </mrow> </msup> <mi>X</mi> </math></EquationSource> <EquationSource Format="TEX">\( \sup _{g \in {\operatorname{Poly}}(\mathbb{Z}\to G)} \left | \sum _{x &lt; n \leq x+H} (\Lambda (n)-\Lambda ^{\sharp }(n)) \overline{F}(g(n) \Gamma ) \right | \ll H\log ^{-A} X \)</EquationSource> </Equation> for any fixed <InlineEquation ID="IEq14"> <EquationSource Format="MATHML"><math> <mi>A</mi> <mo>&gt;</mo> <mn>0</mn> </math></EquationSource> <EquationSource Format="TEX">$A&gt;0$</EquationSource> </InlineEquation>, and that when <InlineEquation ID="IEq15"> <EquationSource Format="MATHML"><math> <msup> <mi>X</mi> <mi>ε</mi> </msup> <mo>≤</mo> <mi>H</mi> <mo>≤</mo> <mi>X</mi> </math></EquationSource> <EquationSource Format="TEX">$X^{\varepsilon } \leq H \leq X$</EquationSource> </InlineEquation> we have, for almost all <InlineEquation ID="IEq16"> <EquationSource Format="MATHML"><math> <mi>x</mi> <mo>∈</mo> <mo stretchy="false">[</mo> <mi>X</mi> <mo>,</mo> <mn>2</mn> <mi>X</mi> <mo stretchy="false">]</mo> </math></EquationSource> <EquationSource Format="TEX">$x \in [X, 2X]$</EquationSource> </InlineEquation>, <Equation ID="Equb"> <EquationSource Format="MATHML"><math> <munder> <mo movablelimits="false">sup</mo> <mrow> <mi>g</mi> <mo>∈</mo> <mo>Poly</mo> <mo stretchy="false">(</mo> <mi mathvariant="double-struck">Z</mi> <mo stretchy="false">→</mo> <mi>G</mi> <mo stretchy="false">)</mo> </mrow> </munder> <mrow> <mo>|</mo> <munder> <mo movablelimits="false">∑</mo> <mrow> <mi>x</mi> <mo>&lt;</mo> <mi>n</mi> <mo>≤</mo> <mi>x</mi> <mo>+</mo> <mi>H</mi> </mrow> </munder> <mo stretchy="false">(</mo> <msub> <mi>d</mi> <mi>k</mi> </msub> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> <mo>−</mo> <msubsup> <mi>d</mi> <mi>k</mi> <mi mathvariant="normal">♯</mi> </msubsup> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> <mo stretchy="false">)</mo> <mover accent="true"> <mi>F</mi> <mo>‾</mo> </mover> <mo stretchy="false">(</mo> <mi>g</mi> <mo stretchy="false">(</mo> <mi>n</mi> <mo stretchy="false">)</mo> <mi mathvariant="normal">Γ</mi> <mo stretchy="false">)</mo> <mo>|</mo> </mrow> <mo>=</mo> <mi>o</mi> <mo stretchy="false">(</mo> <mi>H</mi> <msup> <mo>log</mo> <mrow> <mi>k</mi> <mo>−</mo> <mn>1</mn> </mrow> </msup> <mi>X</mi> <mo stretchy="false">)</mo> <mo>.</mo> </math></EquationSource> <EquationSource Format="TEX">\( \sup _{g \in {\operatorname{Poly}}(\mathbb{Z}\to G)} \left | \sum _{x &lt; n \leq x+H} (d_{k}(n)-d_{k}^{\sharp }(n)) \overline{F}(g(n)\Gamma ) \right | = o(H \log ^{k-1} X). \)</EquationSource> </Equation> As a consequence, we show that the short interval Gowers norms <InlineEquation ID="IEq17"> <EquationSource Format="MATHML"><math> <msub> <mrow> <mo stretchy="false">∥</mo> <mi mathvariant="normal">Λ</mi> <mo>−</mo> <msup> <mi mathvariant="normal">Λ</mi> <mi mathvariant="normal">♯</mi> </msup> <mo stretchy="false">∥</mo> </mrow> <mrow> <msup> <mi>U</mi> <mi>s</mi> </msup> <mo stretchy="false">(</mo> <mi>X</mi> <mo>,</mo> <mi>X</mi> <mo>+</mo> <mi>H</mi> <mo stretchy="false">]</mo> </mrow> </msub> </math></EquationSource> <EquationSource Format="TEX">$\|\Lambda -\Lambda ^{\sharp }\|_{U^{s}(X,X+H]}$</EquationSource> </InlineEquation> and <InlineEquation ID="IEq18"> <EquationSource Format="MATHML"><math> <msub> <mrow> <mo stretchy="false">∥</mo> <msub> <mi>d</mi> <mi>k</mi> </msub> <mo>−</mo> <msubsup> <mi>d</mi> <mi>k</mi> <mi mathvariant="normal">♯</mi> </msubsup> <mo stretchy="false">∥</mo> </mrow> <mrow> <msup> <mi>U</mi> <mi>s</mi> </msup> <mo stretchy="false">(</mo> <mi>X</mi> <mo>,</mo> <mi>X</mi> <mo>+</mo> <mi>H</mi> <mo stretchy="false">]</mo> </mrow> </msub> </math></EquationSource> <EquationSource Format="TEX">$\|d_{k}-d_{k}^{\sharp }\|_{U^{s}(X,X+H]}$</EquationSource> </InlineEquation> are also asymptotically small for any fixed <InlineEquation ID="IEq19"> <EquationSource Format="MATHML"><math> <mi>s</mi> </math></EquationSource> <EquationSource Format="TEX">$s$</EquationSource> </InlineEquation> in the same ranges of <InlineEquation ID="IEq20"> <EquationSource Format="MATHML"><math> <mi>H</mi> </math></EquationSource> <EquationSource Format="TEX">$H$</EquationSource> </InlineEquation>. This in turn allows us to establish the Hardy–Littlewood conjecture and the divisor correlation conjecture with a short average over one variable. Our main new ingredients are type <InlineEquation ID="IEq21"> <EquationSource Format="MATHML"><math> <mi mathvariant="italic">II</mi> </math></EquationSource> <EquationSource Format="TEX">$\mathit{II}$</EquationSource> </InlineEquation> estimates obtained by developing a “contagion lemma” for nilsequences and then using this to “scale up” an approximate functional equation for the nilsequence to a larger scale. This extends an approach developed by Walsh for Fourier uniformity.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Higher uniformity of arithmetic functions in short intervals II. Almost all intervals

  • Kaisa Matomäki,
  • Maksym Radziwiłł,
  • Xuancheng Shao,
  • Terence Tao,
  • Joni Teräväinen

摘要

We study higher uniformity properties of the von Mangoldt function Λ $\Lambda $ , the Möbius function μ $\mu $ , and the divisor functions d k $d_{k}$ on short intervals ( x , x + H ] $(x,x+H]$ for almost all x [ X , 2 X ] $x \in [X, 2X]$ . Let Λ $\Lambda ^{\sharp }$ and d k $d_{k}^{\sharp }$ be suitable approximants of Λ $\Lambda $ and d k $d_{k}$ , G / Γ $G/\Gamma $ a filtered nilmanifold, and F : G / Γ C $F \colon G/\Gamma \to \mathbb{C}$ a Lipschitz function. Then our results imply for instance that when X 1 / 3 + ε H X $X^{1/3+\varepsilon } \leq H \leq X$ we have, for almost all x [ X , 2 X ] $x \in [X, 2X]$ , sup g Poly ( Z G ) | x < n x + H ( Λ ( n ) Λ ( n ) ) F ( g ( n ) Γ ) | H log A X \( \sup _{g \in {\operatorname{Poly}}(\mathbb{Z}\to G)} \left | \sum _{x < n \leq x+H} (\Lambda (n)-\Lambda ^{\sharp }(n)) \overline{F}(g(n) \Gamma ) \right | \ll H\log ^{-A} X \) for any fixed A > 0 $A>0$ , and that when X ε H X $X^{\varepsilon } \leq H \leq X$ we have, for almost all x [ X , 2 X ] $x \in [X, 2X]$ , sup g Poly ( Z G ) | x < n x + H ( d k ( n ) d k ( n ) ) F ( g ( n ) Γ ) | = o ( H log k 1 X ) . \( \sup _{g \in {\operatorname{Poly}}(\mathbb{Z}\to G)} \left | \sum _{x < n \leq x+H} (d_{k}(n)-d_{k}^{\sharp }(n)) \overline{F}(g(n)\Gamma ) \right | = o(H \log ^{k-1} X). \) As a consequence, we show that the short interval Gowers norms Λ Λ U s ( X , X + H ] $\|\Lambda -\Lambda ^{\sharp }\|_{U^{s}(X,X+H]}$ and d k d k U s ( X , X + H ] $\|d_{k}-d_{k}^{\sharp }\|_{U^{s}(X,X+H]}$ are also asymptotically small for any fixed s $s$ in the same ranges of H $H$ . This in turn allows us to establish the Hardy–Littlewood conjecture and the divisor correlation conjecture with a short average over one variable. Our main new ingredients are type II $\mathit{II}$ estimates obtained by developing a “contagion lemma” for nilsequences and then using this to “scale up” an approximate functional equation for the nilsequence to a larger scale. This extends an approach developed by Walsh for Fourier uniformity.