https://arxiv.org/abs/2104.14548
With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representations (Debidatta Dwibedi, Yusuf Aytar, Jonathan Tompson, Pierre Sermanet, Andrew Zisserman)
augment된 이미지가 아니라 nearest neighbor를 positive pair로 사용하는 self supervised learning. 성능이 꽤 올라오는 동시에 흥미로운 방향의 접근 같네요.
#contrastive_learning #knn