# [PAPER] A Simple Framework for contrastive learning of visual representations ![](https://hackmd.io/_uploads/HkOVfqcFh.png) :::info **Author** : Ting Chen, Simon Kornblith, Mohammad Norouzi, Geoffrey Hinton Proceedings of the 37th International Conference on Machine Learning **Paper Link** : http://proceedings.mlr.press/v119/chen20j.html **Code** : https://github.com/google-research/simclr ::: ## Content * This paper present SimCLR:a simple framework for contrastive learning of visual representations. The authors systematically study the major components of their framework and show how it outperforms previous methods for self-supervised and semi-supervised learning on ImageNet. They also provide a comprehensive comparison of their design choices with those of previous work. The strength of this simple framework suggests that self-supervised learning remains undervalued.