It is shown that FAL provides complementary benefits to Confidence-Aware Learning (CAL) using temporal instability as a regularization signal yielding a more self-aware model, and that combining FAL and CAL with focal loss yields the strongest calibration in this study.
Awanish Kumar, Soumyadeep Ghosh, Akshita Sharma et al.· Proceedings of the 32nd ACM...· 0 citations
Modern neural networks are often miscalibrated, assigning high confidence to predictions they have not learned stably, which leads to overconfident errors under noise, imbalance, and distribution shift. We propose Forgetting-Aware Learning (FAL) using a simple ranking regularizer that leverages forgetting events during...
Awanish Kumar, Soumyadeep Ghosh, Akshita Sharma et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.