The coordinate projections of a Euclidean product are nonexpansive and the quadratic cost function is continuous; consequently a bound on the quadratic cost that holds along a weakly convergent sequence of probability measures holds in the limit.
In the setting of Probability Measures on Euclidean Space and Random Vectors: Standing Notation, let satisfy and let denote the Euclidean distance, on and on as the dimension of its arguments requires, so that by claim 2 of Properties of the Euclidean Norm on : Square, Distance, Vanishing, Coordinate Bounds, Homogeneity, Triangle Inequalities and Expansion of a Square. Let carry the absolute-value metric, and let continuity and the Lipschitz property be those of the metric spaces so named. As in Probability Measures on Euclidean Space and Random Vectors: Standing Notation (v), are the coordinate projections for the splitting .
Let be the map with , which is Borel by Pairs of Euclidean Points: Coordinate Projections, Pairings, the Product Measure on a Euclidean Space, Borel Norm Functions and Finite Sets (iv) and nonnegative by Nonnegativity of Squares in an Ordered Field, being a square, and for put
the integral of Probability Measures on Euclidean Space and Random Vectors: Standing Notation (iii) of read as a map into . For and a coupling , is the quadratic cost .
Then the following hold.
1. (The coordinate projections are nonexpansive) Let satisfy and . Then the coordinate projections and of Pairs of Euclidean Points: Coordinate Projections, Pairings, the Product Measure on a Euclidean Space, Borel Norm Functions and Finite Sets (i) are Lipschitz with constant , and continuous on .
2. (The cost function is continuous) The map with value at is Lipschitz with constant , and is continuous on .
3. (A cost bound passes to a weak limit) Let be a sequence in that converges weakly to on the metric space , and let be such that for every with there is with
Then ; in particular is a real number.
Loading…
No relations recorded yet.