Showing 2026Show all
2 papers · 1 filter
stat.ML2026
rSDNet: Unified Robust Neural Learning against Label Noise and Adversarial Attacks
Suryasis Jana, Abhik Ghosh
Neural networks are central to modern artificial intelligence, yet their training remains highly sensitive to data contamination. Standard neural classifiers are trained by minimiz…
stat.ML2026
Provably robust learning of regression neural networks using -divergences
Abhik Ghosh, Suryasis Jana
Regression neural networks (NNs) are most commonly trained by minimizing the mean squared prediction error, which is highly sensitive to outliers and data contamination. Existing r…