Improving Contrastive Learning on Imbalanced Seed Data via Open-World Sampling
Open-world sampling strategy for balanced contrastive learning
#contrastive-learning
#data-sampling
#long-tail
On Model Calibration for Long-Tailed Object Detection and Instance Segmentation
Model-agnostic recalibration of class imbalance for computer vision tasks
#long-tail
#object-detection
#instance-segmentation
Disentangling Sampling and Labeling Bias for Learning in Large-Output Spaces
Negative sampling induces sampling bias, but can help on long-tail dataset
#negative-sampling
#long-tail