3 papers
cs.LG2024
Approximation and generalization properties of the random projection classification method
Mireille Boutin, Evzenie Coupkova
The generalization gap of a classifier is related to the complexity of the set of functions among which the classifier is chosen. We study a family of low-complexity classifiers co…
stat.ML2024
A new model for natural groupings in high-dimensional data
Mireille Boutin, Evzenie Coupkova
Clustering aims to divide a set of points into groups. The current paradigm assumes that the grouping is well-defined (unique) given the probability model from which the data is dr…
cs.LG2024
On the Rashomon ratio of infinite hypothesis sets
Evzenie Coupkova, Mireille Boutin
Given a classification problem and a family of classifiers, the Rashomon ratio measures the proportion of classifiers that yield less than a given loss. Previous work has explored…