BFGS with Update Skipping and Varying Memory

TitleBFGS with Update Skipping and Varying Memory
Publication TypeJournal Articles
Year of Publication1998
AuthorsKolda TG, O'Leary DP, Nazareth L
JournalSIAM Journal on Optimization
Volume8
Issue4
Pagination1060 - 1083
Date Published1998///
Keywordsbfgs, Broyden family, limited-memory, minimization, quasi-Newton, update skipping
Abstract

We give conditions under which limited-memory quasi-Newton methods with exact line searches will terminate in n steps when minimizing n-dimensional quadratic functions. We show that although all Broyden family methods terminate in n steps in their full-memory versions, only BFGS does so with limited-memory. Additionally, we show that full-memory Broyden family methods with exact line searches terminate in at most n + p steps when p matrix updates are skipped. We introduce new limited-memory BFGS variants and test them on nonquadratic minimization problems.

URLhttp://link.aip.org/link/?SJE/8/1060/1
DOI10.1137/S1052623496306450