How to Harden AI Models Against Poisoning Attacks
Techniques include robust training, anomaly detection in datasets, and secure aggregation.
data poisoning
Data Poisoning Attacks in ML Security
Data validation, provenance, and robust aggregation as defenses.