A Simple Log-based Loss Function for Ordinal Text Classification

1 code implementation • COLING 2022 • François Castagnos, Martin Mihelich, Charles Dognin

The cross-entropy loss function is widely used and generally considered the default loss function for text classification.

text-classification Text Classification

A Characterization of Mean Squared Error for Estimator with Bagging

no code implementations • 7 Aug 2019 • Martin Mihelich, Charles Dognin, Yan Shu, Michael Blot

First, we prove that for any estimator, increasing the number of bagged estimators $N$ in the average can only reduce the MSE.