Establishes optimal dimension-free sampling complexities (k²/ε² for L2/k, k/ε² for L1/k, linear in k under bounded derivative) for regularized classification losses with matching lower bounds.
We will use s(a) ≥ ∥a∥2 2 + 2, which is squared norm sampling and can be handled analogously to the previous norm sampling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Optimal Dimension-Free Sampling for Regularized Classification
Establishes optimal dimension-free sampling complexities (k²/ε² for L2/k, k/ε² for L1/k, linear in k under bounded derivative) for regularized classification losses with matching lower bounds.