TY - JOUR
T1 - On the use of polynomial models in multiobjective directional direct search
AU - Brás, Carmo P.
AU - Custódio, Ana Luísa
N1 - FCT - Fundacao para a Ciencia e a Tecnologia PTDC/MAT-APL/28400/2017;
UIDB/00297/2020.
PY - 2020/10/12
Y1 - 2020/10/12
N2 - Polynomial interpolation or regression models are an important tool in Derivative-free Optimization, acting as surrogates of the real function. In this work, we propose the use of these models in the multiobjective framework of directional direct search, namely the one of Direct Multisearch. Previously evaluated points are used to build quadratic polynomial models, which are minimized in an attempt of generating nondominated points of the true function, defining a search step for the algorithm. Numerical results state the competitiveness of the proposed approach.
AB - Polynomial interpolation or regression models are an important tool in Derivative-free Optimization, acting as surrogates of the real function. In this work, we propose the use of these models in the multiobjective framework of directional direct search, namely the one of Direct Multisearch. Previously evaluated points are used to build quadratic polynomial models, which are minimized in an attempt of generating nondominated points of the true function, defining a search step for the algorithm. Numerical results state the competitiveness of the proposed approach.
KW - Derivative-free optimization
KW - Direct search methods
KW - Minimum Frobenius norm models
KW - Multiobjective optimization
KW - Quadratic polynomial interpolation and regression
UR - http://www.scopus.com/inward/record.url?scp=85092490009&partnerID=8YFLogxK
U2 - 10.1007/s10589-020-00233-8
DO - 10.1007/s10589-020-00233-8
M3 - Article
AN - SCOPUS:85092490009
SN - 0926-6003
VL - 77
SP - 897
EP - 918
JO - Computational Optimization And Applications
JF - Computational Optimization And Applications
ER -