In a sharp-floor Gaussian regime, subtracting a same-sample floor estimate from retained PCR denominators makes the conditional prediction risk of dPCR vanish relative to the best ordinary PCR rank.
Roy Frostig, Cameron Musco, Christopher Musco, and Aaron Sidford
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.ST 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
De-floored Principal Component Regression: When Rank Selection Alone Is Insufficient for Prediction
In a sharp-floor Gaussian regime, subtracting a same-sample floor estimate from retained PCR denominators makes the conditional prediction risk of dPCR vanish relative to the best ordinary PCR rank.