TheoremBase

Integration by Parts for the Gibbs Measure of a Gibbs-Sobolev Potential, and Closability of the Noise Gradient in Its Square-Integrable Space

For a Gibbs-Sobolev potential the Gibbs measure has finite second moments, satisfies the Gaussian integration by parts formula with the extra drift given by the partial derivatives of the potential over the temperature, and the noise gradient on cylindrical functions is closable in its square-integrable space.

Statement

In the settings of The Real Numbers: Standing Notation and Background and Gaussian Analysis Relative to a Diagonal Gaussian Reference Measure with Noise Weights: Standing Notation, let β,q∈R\beta,q\in\mathbb{R} satisfy 0<β0<\beta and 1<q1<q, and let VV be a Gibbs-Sobolev potential at temperature β\beta with exponent qq, with partial derivatives ∂kV\partial_{k}V (k∈Nk\in\mathbb{N}). FCb1(X)\mathcal{F}C^{1}_{b}(X), the partial derivatives ∂kφ\partial_{k}\varphi and the noise gradients ∇aφ\nabla_{a}\varphi of its members are those of Gaussian Analysis Relative to a Diagonal Gaussian Reference Measure with Noise Weights: Standing Notation §cylindrical.

1. (The Gibbs measure) The function exp⁡(−V/β)\exp(-V/\beta), with exp⁡\exp the exponential function, is Borel and integrable with respect to γc\gamma_{c}, so the Gibbs measure γβV\gamma^{V}_{\beta} of VV at temperature β\beta is defined; and γβV∈P2(X)\gamma^{V}_{\beta}\in\mathcal{P}_{2}(X).

2. (Square-integrability) For every k∈Nk\in\mathbb{N} the coordinate x↦xkx\mapsto x_{k} and ∂kV\partial_{k}V are square-integrable with respect to γβV\gamma^{V}_{\beta}.

3. (Integration by parts) For every φ∈FCb1(X)\varphi\in\mathcal{F}C^{1}_{b}(X) and k∈Nk\in\mathbb{N}, the function x↦φ(x)(xk/ck+∂kV(x)/β)x\mapsto\varphi(x)\bigl(x_{k}/c_{k}+\partial_{k}V(x)/\beta\bigr) is integrable with respect to γβV\gamma^{V}_{\beta}, and

∫X∂kφ dγβV=∫Xφ(x)(xkck+∂kV(x)β)γβV(dx).\int_{X}\partial_{k}\varphi\,d\gamma^{V}_{\beta}=\int_{X}\varphi(x)\Bigl(\frac{x_{k}}{c_{k}}+\frac{\partial_{k}V(x)}{\beta}\Bigr)\gamma^{V}_{\beta}(dx).

4. (Closability) Let L2(γβV;Xa)L^{2}(\gamma^{V}_{\beta};X^{a}) be the real Hilbert space of Probability Measures on a Hilbert Space Transported in the Noise Norm: Standing Notation §fields for the measure γβV\gamma^{V}_{\beta}, with norm ∥⋅∥γβV\lVert\cdot\rVert_{\gamma^{V}_{\beta}}. If φj∈FCb1(X)\varphi_{j}\in\mathcal{F}C^{1}_{b}(X) (j∈Nj\in\mathbb{N}) and G∈L2(γβV;Xa)G\in L^{2}(\gamma^{V}_{\beta};X^{a}) satisfy ∫Xφj2 dγβV→0\int_{X}\varphi_{j}^{2}\,d\gamma^{V}_{\beta}\to0 and ∥∇aφj−G∥γβV→0\lVert\nabla_{a}\varphi_{j}-G\rVert_{\gamma^{V}_{\beta}}\to0 as j→∞j\to\infty, then G=0G=0.

Proofs

Log in to submit a proof.

Loading...

Citations

Loading…

Dependencies

Loading…

Related

0 relations

Curated associations between results. These are editable and subjective — they do not replace the dependency graph, which is derived from the references in the text.

No relations recorded yet.

Comments

Log in to comment.

Loading…