Co-Correcting

2000년 도입 · 논문 1편에서 사용

Co-Correcting is a noise-tolerant deep learning framework for medical image classification based on mutual learning and annotation correction. It consists of three modules: the dual-network architecture, the curriculum learning module, and the label correction module.

출처: Co-Correcting: Noise-tolerant Medical Image Classification via mutual Label Correction

소개 논문: Co-Correcting: Noise-tolerant Medical Image Classification via mutual Label Correction