Expected Decrease for Derivative-free Algorithms Using Random Subspaces

Abstract: Derivative-free algorithms seek the minimum of a given function based only on function values queried at appropriate points. Their performance is known to worsen as the problem dimension increases. Recent advances in developing randomized derivative-free techniques have tackled this issue by working in low-dimensional subspaces that are drawn at random in an iterative fashion. In this talk, we present analysis for derivative-free algorithms that employing random subspaces to obtain understanding of the expected decrease achieved per function evaluation.

Mathematics

Audience: researchers in the topic

PIMS-CORDS SFU Operations Research Seminar