Proposes a robust sequential kernel conditional independence test via adaptive betting and truncate-and-shift calibration that reduces Type I error inflation from Model-X estimation errors while retaining power.
Title resolution pending
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
-
Sequential Kernel-based Conditional Independence Testing via Adaptive Betting
Proposes a robust sequential kernel conditional independence test via adaptive betting and truncate-and-shift calibration that reduces Type I error inflation from Model-X estimation errors while retaining power.