A random shift of Gaussian inputs forces the first Hermite coefficient of any non-linear target to be large, yielding near-linear sample complexity independent of the target's information exponent, and a similar result holds for juntas.
The Distribution of Values of Analytic Functions on Convex Bodies
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Proceeding the study of local properties of analytic functions started in [Br] we prove new dimensionless inequalities for such functions in terms of their Chebyshev degree. As a consequence, we obtain the reverse Holder inequalities for analytic functions with absolute (i.e., independent of dimension) constants. For polynomials such inequalities were recently proved by Bobkov who sharpened and generalized the previous Bourgain result and by Sodin and Volberg.
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Low-dimensional Functions are Efficiently Learnable under Randomly Biased Distributions
A random shift of Gaussian inputs forces the first Hermite coefficient of any non-linear target to be large, yielding near-linear sample complexity independent of the target's information exponent, and a similar result holds for juntas.