An adaptive stopping algorithm that returns a high-probability upper bound on the best disparate-impact improvement still attainable by continued model retraining.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Statistical Guarantees in the Search for Less Discriminatory Algorithms
An adaptive stopping algorithm that returns a high-probability upper bound on the best disparate-impact improvement still attainable by continued model retraining.