Incremental space-filling design based on coverings and spacings: improving upon low discrepancy sequences
Résumé
The paper addresses the problem of defining families of ordered sequences $\{\xb_i\}_{i\in \mathds{N}}$ of elements of a compact subset $\SX$ of $\mathds{R}^d$ whose prefixes $\Xb_n=\{\xb_i\}_{i=1}^{n}$, for all orders $n$, have good space-filling properties as measured by the dispersion (covering radius) criterion. Our ultimate aim is the definition of incremental algorithms that generate sequences $\Xb_n$ with small optimality gap, i.e., with a small increase in the maximum distance between points of $\SX$ and the elements of $\Xb_n$ with respect to the optimal solution $\Xb_n^\star$. The paper is a first step in this direction, presenting incremental design algorithms with proven optimality bound for one-parameter families of criteria based on coverings and spacings that both converge to dispersion for large values of their parameter. The examples presented show that the covering-based method outperforms state-of-the-art competitors, including coffee-house, suggesting that it inherits from its guaranteed 50\% optimality gap.
Origine : Fichiers produits par l'(les) auteur(s)