Every class with Gaussian surface area at most Gamma admits degree O-tilde(Gamma squared over epsilon squared) non-negative L1-approximating polynomials for its indicators under the standard Gaussian.
Distribution-Independent Reliable Learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Note on Non-Negative $L_1$-Approximating Polynomials
Every class with Gaussian surface area at most Gamma admits degree O-tilde(Gamma squared over epsilon squared) non-negative L1-approximating polynomials for its indicators under the standard Gaussian.