Calibration of Few-Shot Classification Tasks: Mitigating Misconfidence from Distribution Mismatch
Sign Up to like & getrecommendations! Published in 2022 at "IEEE Access"
DOI: 10.1109/access.2022.3176090
Abstract: As many meta-learning algorithms improve performance in solving few-shot classification problems for practical applications, the accurate prediction of uncertainty is considered essential. In meta-training, the algorithm treats all generated tasks equally and updates the model… read more here.
Keywords: classification; shot classification; distribution mismatch; model ... See more keywords