Skip to content

Latest commit

 

History

History
8 lines (5 loc) · 459 Bytes

201116 AdCo.md

File metadata and controls

8 lines (5 loc) · 459 Bytes

https://arxiv.org/abs/2011.08435

AdCo: Adversarial Contrast for Efficient Learning of Unsupervised Representations from Self-Trained Negative Adversaries (Qianjiang Hu, Xiao Wang, Wei Hu, Guo-Jun Qi)

contrastive learning 알고리즘이 또다시 등장! negative 샘플 임베딩 셋을 두고 adversarial하게 negative 샘플을 업데이트해나가는 방식. 연산 소모량에서 이점이 있는 듯.

#contrastive_learning #adversarial_training