Implemention of Dimensionality reduction by learning an invariant mapping in CVPR2006

• In the abstract of MOCO, The DrLIM was introduced, which utilizes the perspective of contrastive learning.

• left expression describe "similar" and right expression describe "dissimilar"
  

• MNIST
  • train: 3000 + a (Augmentation of -6, -3, and 3 degree rotation plus 6 pixel dropout.) of each class
  • test: 400 of each class
• training set
  • ex) img1, img2, label. img1, img2 is random shuffle of "4" and "9". and label is write under the description.
• Using class of "4" and "9"
  • label 0: same class image
  • label 1: different class image