<p>Rademacher's Theorem can be interpreted as an almost everywhere little-o improvement principle: if a function admits a uniform pointwise first-order Lipschitz control at every point, then this control improves to a vanishing one at almost every point. In the language of Calderón-Zygmund pointwise spaces, this means that<Equation ID="Equa"> <EquationSource Format="TEX">\(f \in T^\infty_1(x) \ \ {\rm for\, all}\,\, x \in \mathbb{R}^d {\implies}\, f \in t^\infty_1(x) \ \ \text{for a.e.}\,\, x \in \mathbb{R}^d.\)</EquationSource> <EquationSource Format="MATHML"><math display="block"> <mrow> <mi>f</mi> <mo>∈</mo> <msubsup> <mi>T</mi> <mn>1</mn> <mi>∞</mi> </msubsup> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mrow> <mspace width="4pt" /> <mspace width="4pt" /> <mrow> <mi mathvariant="normal">for</mi> <mspace width="0.166667em" /> <mi mathvariant="normal">all</mi> </mrow> <mspace width="0.166667em" /> <mspace width="0.166667em" /> <mi>x</mi> <mo>∈</mo> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> <mo stretchy="false">⇒</mo> <mspace width="0.166667em" /> <mi>f</mi> <mo>∈</mo> <msubsup> <mi>t</mi> <mn>1</mn> <mi>∞</mi> </msubsup> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mrow> <mspace width="4pt" /> <mspace width="4pt" /> <mtext>for a.e.</mtext> <mspace width="0.166667em" /> <mspace width="0.166667em" /> <mi>x</mi> <mo>∈</mo> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> <mo>.</mo> </mrow> </math></EquationSource> </Equation></p><p>The purpose of this paper is to establish an analogous almost everywhere improvement principle in a refined <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(L^p\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mi>L</mi> <mi>p</mi> </msup> </math></EquationSource> </InlineEquation> setting. We consider pointwise Calderón-Zygmund spaces <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(T^p_{\phi}(x)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msubsup> <mi>T</mi> <mi>ϕ</mi> <mi>p</mi> </msubsup> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> defined via polynomial approximation in <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(L^p\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mi>L</mi> <mi>p</mi> </msup> </math></EquationSource> </InlineEquation> with a function parameter <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(\phi\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>ϕ</mi> </math></EquationSource> </InlineEquation>, allowing for fractional regularity indices and logarithmic corrections through Boyd functions. We prove that, under natural assumptions on <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(\phi\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>ϕ</mi> </math></EquationSource> </InlineEquation>, the uniform membership<Equation ID="Equb"> <EquationSource Format="TEX">\(f \in T^p_{\phi}(x) \quad {\rm for\,\, all}\,\, x \in E\)</EquationSource> <EquationSource Format="MATHML"><math display="block"> <mrow> <mi>f</mi> <mo>∈</mo> <msubsup> <mi>T</mi> <mi>ϕ</mi> <mi>p</mi> </msubsup> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mrow> <mspace width="1em" /> <mrow> <mi mathvariant="normal">for</mi> <mspace width="0.166667em" /> <mspace width="0.166667em" /> <mi mathvariant="normal">all</mi> </mrow> <mspace width="0.166667em" /> <mspace width="0.166667em" /> <mi>x</mi> <mo>∈</mo> <mi>E</mi> </mrow> </math></EquationSource> </Equation>on a measurable set <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(E \subset \mathbb{R}^d\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>E</mi> <mo>⊂</mo> <msup> <mrow> <mi mathvariant="double-struck">R</mi> </mrow> <mi>d</mi> </msup> </mrow> </math></EquationSource> </InlineEquation> implies an almost everywhere improvement to a vanishing approximation rate, namely<Equation ID="Equc"> <EquationSource Format="TEX">\(f \in t^p_{\phi,n+1}(x) \quad \text{for a.e. } x \in E,\)</EquationSource> <EquationSource Format="MATHML"><math display="block"> <mrow> <mi>f</mi> <mo>∈</mo> <msubsup> <mi>t</mi> <mrow> <mi>ϕ</mi> <mo>,</mo> <mi>n</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>p</mi> </msubsup> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mrow> <mspace width="1em" /> <mtext>for a.e.</mtext> <mspace width="0.333333em" /> <mi>x</mi> <mo>∈</mo> <mi>E</mi> <mo>,</mo> </mrow> </math></EquationSource> </Equation>where <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(n &lt; \underline{b}(\phi) \leq \overline{b}(\phi) &lt; n+1\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>n</mi> <mo>&lt;</mo> <munder> <mi>b</mi> <mo>̲</mo> </munder> <mrow> <mo stretchy="false">(</mo> <mi>ϕ</mi> <mo stretchy="false">)</mo> </mrow> <mo>≤</mo> <mover> <mi>b</mi> <mo>¯</mo> </mover> <mrow> <mo stretchy="false">(</mo> <mi>ϕ</mi> <mo stretchy="false">)</mo> </mrow> <mo>&lt;</mo> <mi>n</mi> <mo>+</mo> <mn>1</mn> </mrow> </math></EquationSource> </InlineEquation>.</p><p>The proof combines measurability arguments, a generalized Whitney extension theorem, and fine properties of Sobolev spaces. We also show that this result is essentially sharp: in general, one cannot expect almost everywhere membership in <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(t^p_{\phi,n}(x)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msubsup> <mi>t</mi> <mrow> <mi>ϕ</mi> <mo>,</mo> <mi>n</mi> </mrow> <mi>p</mi> </msubsup> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> for fractional indices, and explicit counterexamples are provided.</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

Rademacher's theorem for Calderón-Zygmund-type spaces

  • T. Lamby

摘要

Rademacher's Theorem can be interpreted as an almost everywhere little-o improvement principle: if a function admits a uniform pointwise first-order Lipschitz control at every point, then this control improves to a vanishing one at almost every point. In the language of Calderón-Zygmund pointwise spaces, this means that \(f \in T^\infty_1(x) \ \ {\rm for\, all}\,\, x \in \mathbb{R}^d {\implies}\, f \in t^\infty_1(x) \ \ \text{for a.e.}\,\, x \in \mathbb{R}^d.\) f T 1 ( x ) for all x R d f t 1 ( x ) for a.e. x R d .

The purpose of this paper is to establish an analogous almost everywhere improvement principle in a refined \(L^p\) L p setting. We consider pointwise Calderón-Zygmund spaces \(T^p_{\phi}(x)\) T ϕ p ( x ) defined via polynomial approximation in \(L^p\) L p with a function parameter \(\phi\) ϕ , allowing for fractional regularity indices and logarithmic corrections through Boyd functions. We prove that, under natural assumptions on \(\phi\) ϕ , the uniform membership \(f \in T^p_{\phi}(x) \quad {\rm for\,\, all}\,\, x \in E\) f T ϕ p ( x ) for all x E on a measurable set \(E \subset \mathbb{R}^d\) E R d implies an almost everywhere improvement to a vanishing approximation rate, namely \(f \in t^p_{\phi,n+1}(x) \quad \text{for a.e. } x \in E,\) f t ϕ , n + 1 p ( x ) for a.e. x E , where \(n < \underline{b}(\phi) \leq \overline{b}(\phi) < n+1\) n < b ̲ ( ϕ ) b ¯ ( ϕ ) < n + 1 .

The proof combines measurability arguments, a generalized Whitney extension theorem, and fine properties of Sobolev spaces. We also show that this result is essentially sharp: in general, one cannot expect almost everywhere membership in \(t^p_{\phi,n}(x)\) t ϕ , n p ( x ) for fractional indices, and explicit counterexamples are provided.