The hessian estimation evolution strategy

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

We present a novel black box optimization algorithm called Hessian Estimation Evolution Strategy. The algorithm updates the covariance matrix of its sampling distribution by directly estimating the curvature of the objective function. This algorithm design is targeted at twice continuously differentiable problems. For this, we extend the cumulative step-size adaptation algorithm of the CMA-ES to mirrored sampling. We demonstrate that our approach to covariance matrix adaptation is efficient by evaluating it on the BBOB/COCO testbed. We also show that the algorithm is surprisingly robust when its core assumption of a twice continuously differentiable objective function is violated. The approach yields a new evolution strategy with competitive performance, and at the same time it also offers an interesting alternative to the usual covariance matrix update mechanism.

OriginalsprogEngelsk
TitelParallel Problem Solving from Nature – PPSN XVI - 16th International Conference, PPSN 2020, Proceedings
RedaktørerThomas Bäck, Mike Preuss, André Deutz, Michael Emmerich, Hao Wang, Carola Doerr, Heike Trautmann
Antal sider13
ForlagSpringer
Publikationsdato2020
Sider597-609
ISBN (Trykt)9783030581114
DOI
StatusUdgivet - 2020
Begivenhed16th International Conference on Parallel Problem Solving from Nature, PPSN 2020 - Leiden, Holland
Varighed: 5 sep. 20209 sep. 2020

Konference

Konference16th International Conference on Parallel Problem Solving from Nature, PPSN 2020
LandHolland
ByLeiden
Periode05/09/202009/09/2020
NavnLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vol/bind12269 LNCS
ISSN0302-9743

ID: 250555502