For ellipsoid-constrained linear regression with covariate shift, the paper proposes a preconditioned least-squares estimator and proves lower and upper bounds, but the claimed exact match uses incompatible constants.
(80) From the above procedure, we have δ≤ 1 2188ψ lnn tr S, γ ∈ [ δ, 1 2188ψ lnn∑ i>˜κλi ] , β = δ 4376ψ˜κγ lnn
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Optimal Algorithms in Linear Regression under Covariate Shift: On the Importance of Precondition
For ellipsoid-constrained linear regression with covariate shift, the paper proposes a preconditioned least-squares estimator and proves lower and upper bounds, but the claimed exact match uses incompatible constants.