From Noisy Prediction to True Label: Noisy Prediction Calibration via Generative Model

HeeSun Bae, Seungjae Shin, Byeonghu Na, JoonHo Jang, Kyungwoo Song, Il-Chul Moon

Abstract: Noisy labels are inevitable yet problematic in machine learning society. It ruins the generalization of a classifier by making the classifier over-fitted to noisy labels. Existing methods on noisy ...