Establishes minimax bounds c K²/(log n)^{3/2} lower and C K/sqrt(log n) upper on the L2 risk for estimating monotone f:{0,1}^d→[0,1] with total influence I(f)≤K via Fourier thresholding.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.ST 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Influence as soft sparsity: Estimation of monotone functions on $\{0,1\}^d$
Establishes minimax bounds c K²/(log n)^{3/2} lower and C K/sqrt(log n) upper on the L2 risk for estimating monotone f:{0,1}^d→[0,1] with total influence I(f)≤K via Fourier thresholding.