← Search

Enrique Nueve

2 accepted papers

2024

Trading off Consistency and Dimensionality of Convex Surrogates for Multiclass Classification

NeurIPS 2024poster

In multiclass classification over $n$ outcomes, we typically optimize some surrogate loss $L: \mathbb{R}^d \times\mathcal{Y} \to \mathbb{R}$ assigning real-valued error to predictions in $\mathbb{R}^d$. In this paradigm, outcomes must be embedded into the reals with dimension $d \approx n$ in order…

Cited by 0SourcePDFScholar