Geodesic
Box-constrained optimization for minimax supervised learning
Cyprien Gilet1 ; Susana Barbosa2 ; Lionel Fillatre1
1 University of Côte d’Azur, CNRS, I3S laboratory, Sophia-Antipolis, France
2 University of Côte d’Azur, CNRS, laboratory IPMC, Sophia-Antipolis, France
ESAIM. Proceedings, Tome 71 (2021), pp. 101-113 Cet article a éte moissonné depuis la source EDP Sciences

Voir la notice de l'article

In this paper, we present the optimization procedure for computing the discrete boxconstrained minimax classifier introduced in [1, 2]. Our approach processes discrete or beforehand discretized features. A box-constrained region defines some bounds for each class proportion independently. The box-constrained minimax classifier is obtained from the computation of the least favorable prior which maximizes the minimum empirical risk of error over the box-constrained region. After studying the discrete empirical Bayes risk over the probabilistic simplex, we consider a projected subgradient algorithm which computes the prior maximizing this concave multivariate piecewise affine function over a polyhedral domain. The convergence of our algorithm is established.
DOI : 10.1051/proc/202171109

Cyprien Gilet 1 ; Susana Barbosa 2 ; Lionel Fillatre 1

1 University of Côte d’Azur, CNRS, I3S laboratory, Sophia-Antipolis, France
2 University of Côte d’Azur, CNRS, laboratory IPMC, Sophia-Antipolis, France 
@article{EP_2021_71_a9,
     author = {Cyprien Gilet and Susana Barbosa and Lionel Fillatre},
     title = {Box-constrained optimization for minimax supervised learning},
     journal = {ESAIM. Proceedings},
     pages = {101--113},
     year = {2021},
     volume = {71},
     doi = {10.1051/proc/202171109},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.1051/proc/202171109/}
}
TY  - JOUR
AU  - Cyprien Gilet
AU  - Susana Barbosa
AU  - Lionel Fillatre
TI  - Box-constrained optimization for minimax supervised learning
JO  - ESAIM. Proceedings
PY  - 2021
SP  - 101
EP  - 113
VL  - 71
UR  - http://geodesic.mathdoc.fr/articles/10.1051/proc/202171109/
DO  - 10.1051/proc/202171109
LA  - en
ID  - EP_2021_71_a9
ER  - 
%0 Journal Article
%A Cyprien Gilet
%A Susana Barbosa
%A Lionel Fillatre
%T Box-constrained optimization for minimax supervised learning
%J ESAIM. Proceedings
%D 2021
%P 101-113
%V 71
%U http://geodesic.mathdoc.fr/articles/10.1051/proc/202171109/
%R 10.1051/proc/202171109
%G en
%F EP_2021_71_a9
Cyprien Gilet; Susana Barbosa; Lionel Fillatre. Box-constrained optimization for minimax supervised learning. ESAIM. Proceedings, Tome 71 (2021), pp. 101-113. doi: 10.1051/proc/202171109

Cité par Sources :