collaborators

5 papers

cs.LG2026

Weighted Score-Oriented Losses for Temporally Localized Event Prediction

Edoardo Legnaro, Sabrina Guastavino, Francesco Marchetti

Operational event-detection systems are rarely assessed by pointwise accuracy alone. In anomaly detection, changepoint detection, and warning systems, the utility of an alarm depen…

cs.LG2025

Multiclass threshold-based classification

Francesco Marchetti, Edoardo Legnaro, Sabrina Guastavino

In this paper, we introduce a threshold-based framework for multiclass classification that generalizes the standard argmax rule. This is done by replacing the probabilistic interpr…

cs.LG2025

The Multiclass Score-Oriented Loss (MultiSOL) on the Simplex

Francesco Marchetti, Edoardo Legnaro, Sabrina Guastavino

In the supervised binary classification setting, score-oriented losses have been introduced with the aim of optimizing a chosen performance metric directly during the training phas…

cs.LG2025

Multiclass threshold-based classification and model evaluation

Edoardo Legnaro, Sabrina Guastavino, Francesco Marchetti

In this paper, we introduce a threshold-based framework for multiclass classification that generalizes the standard argmax rule. This is done by replacing the probabilistic interpr…

math.NA2025

Feature Understanding and Sparsity Enhancement via 2-Layered kernel machines (2L-FUSE)

Fabiana Camattari, Sabrina Guastavino, Francesco Marchetti +1

We propose a novel sparsity enhancement strategy for regression tasks, based on learning a data-adaptive kernel metric, i.e., a shape matrix, through 2-Layered kernel machines. The…