Data di Pubblicazione:
2024
Abstract:
In this paper we propose a variant of the linear least squares model allowing
practitioners to partition the input features into groups of variables that
they require to contribute similarly to the final result. The output allows
practitioners to assess the importance of each group and of each variable in
the group. We formally show that the new formulation is not convex and provide
two alternative methods to deal with the problem: one non-exact method based on
an alternating least squares approach; and one exact method based on a
reformulation of the problem using an exponential number of sub-problems whose
minimum is guaranteed to be the optimal solution. We formally show the
correctness of the exact method and also compare the two solutions showing that
the exact solution provides better results in a fraction of the time required
by the alternating least squares solution (assuming that the number of
partitions is small). For the sake of completeness, we also provide an
alternative branch and bound algorithm that can be used in place of the exact
method when the number of partitions is too large, and a proof of
NP-completeness of the optimization problem introduced in this paper.
Tipologia CRIS:
03A-Articolo su Rivista
Keywords:
Computer Science - Learning; Computer Science - Learning; Statistics - Machine Learning
Elenco autori:
Roberto Esposito; Mattia Cerrato; Marco Locatelli
Link alla scheda completa:
Link al Full Text:
Pubblicato in: