Laplace on real amounts →
Compare ldp_laplace_real(amount_usd, …) against the integer-bucket pipeline at the same ε. Which estimates the mean better?
Data pre-processed with ε=2.0
Click "Fetch Data" to load the pre-anonymized distribution
Laplace on real amounts →
Compare ldp_laplace_real(amount_usd, …) against the integer-bucket pipeline at the same ε. Which estimates the mean better?