Long-tail learning with rebalanced contrastive loss

Integrating supervised contrastive loss to cross entropy-based classification has recently been proposed as a solution to address the long-tail learning problem. However, when the class imbalance ratio is high, it requires adjusting the supervised contrastive loss to support the tail classes, as the conventional contrastive learning is biased towards head classes by default. To this end, we present Rebalanced Contrastive Learning (RCL), an efficient means to increase the long-tail classification accuracy by addressing three main aspects: 1. Feature space balancedness – Equal division of the feature space among all the classes 2. Intra-Class compactness – Reducing the distance between same-class embeddings 3. Regularization – Enforcing larger margins for tail classes to reduce overfitting. RCL adopts class frequency-based SoftMax loss balancing to supervised contrastive learning loss and exploits scalar multiplied features fed to the contrastive learning loss to enforce compactness. We implement RCL on the Balanced Contrastive Learning (BCL) Framework, which has the SOTA performance. Our experiments on three benchmark datasets CIFAR10-LT,CIFAR100-LT and ImageNet-LT demonstrate the richness of the learnt embeddings and increased top-1 balanced accuracy RCL provides to the BCL framework. We further demonstrate that the performance of RCL as a standalone loss also achieves state-of-the-art level accuracy. • Rebalances supervised contrastive learning to support long-tail classification. • Optimizes the learnt feature distribution to support rare class classification. • Improved performance over datasets: CIFAR10 Lt, CIFAR100 Lt, and ImageNet Lt. • Enhanced class separability, feature space balancedness and intra-class compactness. • Can apply complementary to existing long-tail classification frameworks.