No One Left Behind: Improving the Worst Categories in Long-Tailed Learning

by   Yingxiao Du, et al.

Unlike the case when using a balanced training dataset, the per-class recall (i.e., accuracy) of neural networks trained with an imbalanced dataset are known to vary a lot from category to category. The convention in long-tailed recognition is to manually split all categories into three subsets and report the average accuracy within each subset. We argue that under such an evaluation setting, some categories are inevitably sacrificed. On one hand, focusing on the average accuracy on a balanced test set incurs little penalty even if some worst performing categories have zero accuracy. On the other hand, classes in the "Few" subset do not necessarily perform worse than those in the "Many" or "Medium" subsets. We therefore advocate to focus more on improving the lowest recall among all categories and the harmonic mean of all recall values. Specifically, we propose a simple plug-in method that is applicable to a wide range of methods. By simply re-training the classifier of an existing pre-trained model with our proposed loss function and using an optional ensemble trick that combines the predictions of the two classifiers, we achieve a more uniform distribution of recall values across categories, which leads to a higher harmonic mean accuracy while the (arithmetic) average accuracy is still high. The effectiveness of our method is justified on widely used benchmark datasets.


Inducing Neural Collapse in Deep Long-tailed Learning

Although deep neural networks achieve tremendous success on various clas...

ACE: Ally Complementary Experts for Solving Long-Tailed Recognition in One-Shot

One-stage long-tailed recognition methods improve the overall performanc...

Nested Collaborative Learning for Long-Tailed Visual Recognition

The networks trained on the long-tailed dataset vary remarkably, despite...

Class Balancing GAN with a Classifier in the Loop

Generative Adversarial Networks (GANs) have swiftly evolved to imitate i...

NCL++: Nested Collaborative Learning for Long-Tailed Visual Recognition

Long-tailed visual recognition has received increasing attention in rece...

Text Classification in the Wild: a Large-scale Long-tailed Name Normalization Dataset

Real-world data usually exhibits a long-tailed distribution,with a few f...

Mutual Exclusive Modulator for Long-Tailed Recognition

The long-tailed recognition (LTR) is the task of learning high-performan...

Please sign up or login with your details

Forgot password? Click here to reset