Large Margin Classifiers: Convex Loss, Low Noise, and Convergence Rates

57 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Supersparse linear integer models for optimized medical scoring systems (2015, background): “We consider linear classification models of the form ŷ = sign(λTx), where λ ⊆ RP+1 represents a vector of coefficients [λ0, λ1, . . . , λP ] T and λ0 represents an intercept term.”
On the Consistency of Multiclass Classification Methods (2007, methodology): “analyzed in the literature [4, 6, 13, 9, 1 ]. In this thesis, we investigate the consistency of 2”
Loss Functions for Binary Class Probability Estimation and Classification: Structure and Applications (2005, methodology): “L(Y | q(X)) , minimization of L creates Fisher consistent estimates q(x) of η(x) if Fisher consistency holds pointwise: argmin q∈[0] [1] E Y ∼Bernoulli(η) L(Y | q) = η , ∀ η ∈ [0] [1].”