Computing D-Optimal solutions for huge-scale linear and quadratic response-surface models

TL;DR

This paper develops practical algorithms for computing D-optimal designs in large-scale, structured response-surface models using optimization techniques like row generation, local search, and relaxation methods.

Contribution

It introduces new algorithmic approaches for D-optimality in large, structured models, combining optimization strategies for improved scalability and practicality.

Findings

Algorithms effectively handle large, structured design matrices.

Row generation and relaxation techniques improve computational efficiency.

Practical solutions for high-dimensional response-surface models.

Abstract

We consider algorithmic approaches to the D-optimality problem for cases where the input design matrix is large and highly structured, in particular implicitly specified as a full quadratic or linear response-surface model in several levels of several factors. Using row generation techniques of mathematical optimization, in the context of discrete local-search and continuous relaxation aimed at branch-and-bound solution, we are able to design practical algorithms.