A private labeled-data prior gives an auditor a calculable probability of catching a platform that manipulates its answers to appear fair, while any public prior can be exploited.
Provable Fairness for Neural Network Models using Formal Verification , 2022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Robust ML Auditing using Prior Knowledge
A private labeled-data prior gives an auditor a calculable probability of catching a platform that manipulates its answers to appear fair, while any public prior can be exploited.