AAAIApr, 2021
视觉表征学习的互惠对比学习
Mutual Contrastive Learning for Visual Representation Learning
Chuanguang Yang, Zhulin An, Linhang Cai, Yongjun Xu
TL;DRMutual Contrastive Learning (MCL) is a powerful method for improving feature representations for visual recognition tasks through the mutual interaction and transfer of contrastive distributions among a cohort of networks, achieved through the use of Interactive Contrastive Learning (ICL) to aggregate cross-network embedding information.