Certifiable Robustness for Nearest Neighbor Classifiers

Austen Z. Fan, Paraschos Koutris

ML models are typically trained using large datasets of high quality. However, training datasets often contain inconsistent or incomplete data. To tackle this issue, one solution i…