2 papers
cs.LG2026
Disagreement-Regularized Importance Sampling for Adversarial Label Corruption
Csongor Horváth, Ida-Maria Sintorn, Prashant Singh
Standard Importance Sampling (IS) collapses under label corruption because high-norm examples, prioritized for variance reduction, are often adversarial outliers. We formalize this…
cs.LG2025
Bayesian Robust Aggregation for Federated Learning
Aleksandr Karakulev, Usama Zafar, Salman Toor +1
Federated Learning enables collaborative training of machine learning models on decentralized data. This scheme, however, is vulnerable to adversarial attacks, when some of the cli…