Appendix: Improving Contrastive Learning on Imbalanced Seed Data via Open-World Sampling

B) Details of the employed hyperparameters. For all fine-tuning, the optimizer is set as SGD with momentum of 0.9 and initial learning rate of 30 following [ When fine-tuning for linear separability performance, we train for 30 epochs and decrease the learning rate by 10 times at epochs 10 and 20. The initial lr is set as 0.02 and employing cosine learning rate decay without