One-pass algorithms achieve Õ(M²/ε) space for regression splits and Õ(1/ε) space for Gini splits with matching Ω lower bounds.
On Learning with Integral Operators
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Empirical integral operators with discontinuous non-negative symmetric kernels converge spectrally to their population versions with explicit rates as sample size grows to infinity.
citing papers explorer
-
Nearly Optimal Bounds for Computing Decision Tree Splits in Data Streams
One-pass algorithms achieve Õ(M²/ε) space for regression splits and Õ(1/ε) space for Gini splits with matching Ω lower bounds.
-
Spectral convergence of empirical integral operators with discontinuous kernels
Empirical integral operators with discontinuous non-negative symmetric kernels converge spectrally to their population versions with explicit rates as sample size grows to infinity.