Bayesian quadrature for $H^1(\mu)$ with Poincar\'e inequality on a compact interval

TL;DR

This paper develops a Bayesian quadrature method for functions in $H^1(u)$ spaces on compact intervals, leveraging Poincare9 inequalities to construct optimal quadrature rules with explicit error bounds.

Contribution

It introduces Poincare9 quadrature, a novel Bayesian quadrature approach based on spectral decomposition linked to Poincare9 inequalities, with explicit formulas and numerical procedures.

Findings

Poincare9 quadrature is asymptotically optimal for uniform distributions.

Nodes are nearly evenly spaced and weights approximate the probability density.

Worst-case error scales as O(n^{-1}) for large n.

Abstract

Motivated by uncertainty quantification of complex systems, we aim at finding quadrature formulas of the form $\int_a^b f(x) d\mu(x) = \sum_{i=1}^n w_i f(x_i)$ where $f$ belongs to $H^1(\mu)$. Here, $\mu$ belongs to a class of continuous probability distributions on $[a, b] \subset \mathbb{R}$ and $\sum_{i=1}^n w_i \delta_{x_i}$ is a discrete probability distribution on $[a, b]$. We show that $H^1(\mu)$ is a reproducing kernel Hilbert space with a continuous kernel $K$, which allows to reformulate the quadrature question as a Bayesian (or kernel) quadrature problem. Although $K$ has not an easy closed form in general, we establish a correspondence between its spectral decomposition and the one associated to Poincar\'e inequalities, whose common eigenfunctions form a $T$-system (Karlin and Studden, 1966). The quadrature problem can then be solved in the finite-dimensional proxy space spanned by the first eigenfunctions. The solution is given by a generalized Gaussian quadrature, which we call Poincar\'e quadrature. We derive several results for the Poincar\'e quadrature weights and the associated worst-case error. When $\mu$ is the uniform distribution, the results are explicit: the Poincar\'e quadrature is equivalent to the midpoint (rectangle) quadrature rule. Its nodes coincide with the zeros of an eigenfunction and the worst-case error scales as $\frac{b-a}{2\sqrt{3}}n^{-1}$ for large $n$. By comparison with known results for $H^1(0,1)$, this shows that the Poincar\'e quadrature is asymptotically optimal. For a general $\mu$, we provide an efficient numerical procedure, based on finite elements and linear programming. Numerical experiments provide useful insights: nodes are nearly evenly spaced, weights are close to the probability density at nodes, and the worst-case error is approximately $O(n^{-1})$ for large $n$.