The classical framework of learning from examples is enhanced by the introduction of hard pointwise constraints, i.e., constraints imposed on a finite set of examples that cannot be violated. Such constraints arise, e.g., when requiring coherent decisions of classifiers acting on different views of the same pattern. It is shown that the optimal solution to the learning problem with hard bilateral and linear pointwise constraints can be obtained as the limit of the sequence of optimal solutions to the related learning problems with soft bilateral and linear pointwise constraints, when the penalty parameter tends to infinity. Numerical examples are presented, where hard linear pointwise constraints combined with soft pointwise constraints induced by supervised examples.
Gnecco, G., Gori, M., Melacci, S., Sanguineti, M. (2016). A constrained machine-learning paradigm. In Workshop Proceedings of the XV AI*IA Conference of the Italian Association for Artificial Intelligence.
A constrained machine-learning paradigm
GORI, MARCO;MELACCI, STEFANO;
2016-01-01
Abstract
The classical framework of learning from examples is enhanced by the introduction of hard pointwise constraints, i.e., constraints imposed on a finite set of examples that cannot be violated. Such constraints arise, e.g., when requiring coherent decisions of classifiers acting on different views of the same pattern. It is shown that the optimal solution to the learning problem with hard bilateral and linear pointwise constraints can be obtained as the limit of the sequence of optimal solutions to the related learning problems with soft bilateral and linear pointwise constraints, when the penalty parameter tends to infinity. Numerical examples are presented, where hard linear pointwise constraints combined with soft pointwise constraints induced by supervised examples.File | Dimensione | Formato | |
---|---|---|---|
MLDM_2016.pdf
non disponibili
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
212.25 kB
Formato
Adobe PDF
|
212.25 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11365/1005560