P-COQS builds differentially private conformal prediction sets by replacing the calibration quantile with a binary-search privatized quantile, yielding approximate coverage with a computable error bound.
Simulation Results With Non-DP Random Forest Model A.1 Effect of ϵCP with non-DP model The parameter settings are as in the non-DP NB model and the model’s accuracy is 81%
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Differentially Private Conformal Prediction via Quantile Binary Search
P-COQS builds differentially private conformal prediction sets by replacing the calibration quantile with a binary-search privatized quantile, yielding approximate coverage with a computable error bound.