A measure has a relative score with respect to the Gibbs measure of an admissible cylindrical potential if, coordinate by coordinate, integration by parts against bounded cylindrical functions holds with the Gibbs drift; the components are unique.
In the setting of A Diagonal Gaussian Reference Measure on the Noise Wasserstein Space, Rescaled Heads and Gaussian Tails: Standing Notation, let be an admissible cylindrical potential with head dimension , with the functions of Admissible Cylindrical Potentials on a Hilbert Space in the Noise Geometry §gradient, let be positive, and let be the Gibbs measure of at temperature . is the set of bounded cylindrical functions and are their partial derivatives, which exist by Bounded C^1 Cylindrical Functions: Linear Structure, the Gradient and the Partial Derivatives, Bounds and Integrability, and Density in the Square-Integrable Functions §gradient. Let , the set of The Second Moment of a Borel Probability Measure on a Hilbert Space and the Probability Measures with Finite Second Moment §space, be such that is integrable with respect to for every ; the notion below is defined only for such . Let be the real Hilbert space of The Lebesgue Space of Square-Integrable Functions is a Real Hilbert Space §hilbert, with the inner product and norm of The Lebesgue Space of Square-Integrable Functions is a Real Hilbert Space §inner-product. For and , the class of belongs to and is again written , and is integrable with respect to , by Bounded C^1 Cylindrical Functions: Linear Structure, the Gradient and the Partial Derivatives, Bounds and Integrability, and Density in the Square-Integrable Functions §square-integrable; the function is integrable with respect to by Bounded C^1 Cylindrical Functions: Linear Structure, the Gradient and the Partial Derivatives, Bounds and Integrability, and Density in the Square-Integrable Functions §coordinate-integrable; hence, being positive by Variance Sequences and Their Truncations §variances, the function is integrable with respect to by Linearity and Monotonicity of the Lebesgue Integral §integrable. Next, is Borel and bounded by Bounded C^1 Cylindrical Functions: Linear Structure, the Gradient and the Partial Derivatives, Bounds and Integrability, and Density in the Square-Integrable Functions §bounded-borel, so there is with for every ; and is Borel by Basic Properties of an Admissible Cylindrical Potential: Continuity, Growth under Noise Translations, Integrability, the Tangent Inequality and Tangency of the Noise Gradient §continuity and integrable with respect to , by hypothesis for and being for . So is Borel with on , and is integrable with respect to by the monotonicity of Linearity and Monotonicity of the Lebesgue Integral §nonnegative, applied with and . Hence the function
is integrable with respect to by Linearity and Monotonicity of the Lebesgue Integral §integrable.
(Relative score) The measure has a relative score with respect to if for every there is with
For each there is at most one such : the difference of two of them is orthogonal to every , by linearity of the inner product, and so is the zero vector by Bounded C^1 Cylindrical Functions: Linear Structure, the Gradient and the Partial Derivatives, Bounds and Integrability, and Density in the Square-Integrable Functions §density. The relative score of with respect to is then the sequence , and is its -th component.
Loading…
No relations recorded yet.