<p>This paper is grounded in the long-standing, purely abstract theory of integration with respect to nonlinear operator-valued measures <InlineEquation ID="IEq1"> <EquationSource Format="MATHML"><math> <mi>μ</mi> <mo>:</mo> <mi mathvariant="script">R</mi> <mo stretchy="false">→</mo> <mi>N</mi> <mo stretchy="false">(</mo> <mi>S</mi> <mo>,</mo> <mi>F</mi> <mo stretchy="false">)</mo> </math></EquationSource> <EquationSource Format="TEX">$\mu :\mathcal{R}\to N(S,F)$</EquationSource> </InlineEquation>, which defines the integral operator <InlineEquation ID="IEq2"> <EquationSource Format="MATHML"><math> <mi>T</mi> <mi>f</mi> <mo>≡</mo> <mo>∫</mo> <mi>f</mi> <mspace width="0.2em" /> <mi mathvariant="normal">d</mi> <mi>μ</mi> </math></EquationSource> <EquationSource Format="TEX">$Tf\equiv \int f\,\mathrm {d}\mu $</EquationSource> </InlineEquation> in full generality. By leveraging Pettis and Bochner Radon–Nikodým-type theorems and imposing natural assumptions on measurability and semivariation, we establish that this abstract integral can be represented by a classical scalar integral with a Carathéodory kernel <InlineEquation ID="IEq3"> <EquationSource Format="MATHML"><math> <mi>φ</mi> <mo>:</mo> <mi>S</mi> <mo>×</mo> <mi>X</mi> <mo stretchy="false">→</mo> <mi>F</mi> </math></EquationSource> <EquationSource Format="TEX">$\varphi :S\times X\to F$</EquationSource> </InlineEquation>. This representation, a core contribution of this work, translates the abstract operator integral into the more familiar form <InlineEquation ID="IEq4"> <EquationSource Format="MATHML"><math> <mi>T</mi> <mi>f</mi> <mo>=</mo> <msub> <mo>∫</mo> <mi>X</mi> </msub> <mi>φ</mi> <mo stretchy="false">(</mo> <mi>f</mi> <mo stretchy="false">(</mo> <mi>y</mi> <mo stretchy="false">)</mo> <mo>,</mo> <mi>y</mi> <mo stretchy="false">)</mo> <mspace width="0.2em" /> <mi>d</mi> <mi>m</mi> <mo stretchy="false">(</mo> <mi>y</mi> <mo stretchy="false">)</mo> </math></EquationSource> <EquationSource Format="TEX">$Tf=\int _{X}\varphi (f(y),y)\,dm(y)$</EquationSource> </InlineEquation>. This approach not only provides a powerful analytical framework but also enables the direct use of standard tools for nonlinear integral operators. Moreover, by introducing a Lipschitz condition on the kernel, we derive the nonexpansive inequality <InlineEquation ID="IEq5"> <EquationSource Format="MATHML"><math> <mo stretchy="false">|</mo> <mo stretchy="false">|</mo> <mi>T</mi> <mi>f</mi> <mo>−</mo> <mi>T</mi> <mi>g</mi> <mo stretchy="false">|</mo> <mo stretchy="false">|</mo> <mo>≤</mo> <mo stretchy="false">|</mo> <mo stretchy="false">|</mo> <mi>L</mi> <mo stretchy="false">|</mo> <msub> <mo stretchy="false">|</mo> <msup> <mi>L</mi> <mi>q</mi> </msup> </msub> <mo stretchy="false">|</mo> <mo stretchy="false">|</mo> <mi>f</mi> <mo>−</mo> <mi>g</mi> <mo stretchy="false">|</mo> <msub> <mo stretchy="false">|</mo> <msup> <mi>L</mi> <mi>p</mi> </msup> </msub> </math></EquationSource> <EquationSource Format="TEX">$||Tf-Tg||\le ||L||_{L^{q}}||f-g||_{L^{p}}$</EquationSource> </InlineEquation>, thereby classifying <i>T</i> as a nonexpansive operator (or a strict contraction when <InlineEquation ID="IEq6"> <EquationSource Format="MATHML"><math> <mo stretchy="false">|</mo> <mo stretchy="false">|</mo> <mi>L</mi> <mo stretchy="false">|</mo> <msub> <mo stretchy="false">|</mo> <msup> <mi>L</mi> <mi>q</mi> </msup> </msub> <mo>&lt;</mo> <mn>1</mn> </math></EquationSource> <EquationSource Format="TEX">$||L||_{L^{q}}&lt;1$</EquationSource> </InlineEquation>). Consequently, this framework enables the direct application of classical fixed-point algorithms, such as Picard iterations for contractions and Krasnoselskii–Mann or Halpern schemes for nonexpansive mappings. This work bridges the gap between the abstract theory of operator-valued measures and the practical application of nonexpansive-type operator theory and algorithms.</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

Nonexpansive-type operators from nonlinear operator-valued measures

  • Tomasz Szczypiński

摘要

This paper is grounded in the long-standing, purely abstract theory of integration with respect to nonlinear operator-valued measures μ : R N ( S , F ) $\mu :\mathcal{R}\to N(S,F)$ , which defines the integral operator T f f d μ $Tf\equiv \int f\,\mathrm {d}\mu $ in full generality. By leveraging Pettis and Bochner Radon–Nikodým-type theorems and imposing natural assumptions on measurability and semivariation, we establish that this abstract integral can be represented by a classical scalar integral with a Carathéodory kernel φ : S × X F $\varphi :S\times X\to F$ . This representation, a core contribution of this work, translates the abstract operator integral into the more familiar form T f = X φ ( f ( y ) , y ) d m ( y ) $Tf=\int _{X}\varphi (f(y),y)\,dm(y)$ . This approach not only provides a powerful analytical framework but also enables the direct use of standard tools for nonlinear integral operators. Moreover, by introducing a Lipschitz condition on the kernel, we derive the nonexpansive inequality | | T f T g | | | | L | | L q | | f g | | L p $||Tf-Tg||\le ||L||_{L^{q}}||f-g||_{L^{p}}$ , thereby classifying T as a nonexpansive operator (or a strict contraction when | | L | | L q < 1 $||L||_{L^{q}}<1$ ). Consequently, this framework enables the direct application of classical fixed-point algorithms, such as Picard iterations for contractions and Krasnoselskii–Mann or Halpern schemes for nonexpansive mappings. This work bridges the gap between the abstract theory of operator-valued measures and the practical application of nonexpansive-type operator theory and algorithms.