… we show our class-balanced loss can be used as a generic loss for vi… loss on ILSVRC 2012.
We believe our study on quantifying the effective number of samples and classbalanced loss …

… between both easy samples and hard samples. Aiming at … of loss functions, namely CEFL
loss and CEFL2 loss, based on cross-entropy loss and focal loss to further increase the loss for …

… Class-balanced loss [15] proposed by Lin et al. calculates the effective number of samples
… We compare the results of our method with that of Focal loss [14], Class-Balanced loss [15], …

… [35] design class-balanced loss (CBLoss), which assigns costs to different … based on the
effective number of samples within the neighboring region rather than the number of samples for …

… Loss: A class-level re-balancing method which adjusts the weights of classes based on effective
numbers … Class-balanced loss based on effective number of samples. In CVPR, 2019. 1, …

… of class-balanced losses on different GBDT models, establishing a valuable benchmark. Our
results demonstrate the potential of class-balanced loss … number of samples into the loss …

… Classbalanced loss based on effective number of samples. In 2019 IEEE/CVF
Conference on Computer Vision and Pattern Recognition (CVPR), 2019. 1, 2, 5, 7, 8 [9] Q. …

… based on a class-balanced dynamically weighted loss function where weights are assigned
based … that determines per-sample loss weights of the training data based on their gradient …

… effective sample space calculation result compared with the other two in large datasets [40].
The effective sample numbers … number of effective samples should grow with the number of …

… In terms of balance in class-wise detection performance, weighting by Effective Number of
Samples had the best result followed by Balanced Cross Entropy and by using Inverse Class …