Representation Norm Amplification for Out-of-Distribution Detection in Long-Tail Learning
This addresses the problem of reliable machine learning for models trained on imbalanced datasets, offering a novel solution to a specific bottleneck in OOD detection.
The paper tackles the challenge of out-of-distribution (OOD) detection in long-tail learning by introducing Representation Norm Amplification (RNA), which decouples OOD detection from in-distribution classification using representation norms, resulting in improvements of up to 9.46% in FPR95 and 6.87% in classification accuracy on benchmark datasets.
Detecting out-of-distribution (OOD) samples is a critical task for reliable machine learning. However, it becomes particularly challenging when the models are trained on long-tailed datasets, as the models often struggle to distinguish tail-class in-distribution samples from OOD samples. We examine the main challenges in this problem by identifying the trade-offs between OOD detection and in-distribution (ID) classification, faced by existing methods. We then introduce our method, called \textit{Representation Norm Amplification} (RNA), which solves this challenge by decoupling the two problems. The main idea is to use the norm of the representation as a new dimension for OOD detection, and to develop a training method that generates a noticeable discrepancy in the representation norm between ID and OOD data, while not perturbing the feature learning for ID classification. Our experiments show that RNA achieves superior performance in both OOD detection and classification compared to the state-of-the-art methods, by 1.70\% and 9.46\% in FPR95 and 2.43\% and 6.87\% in classification accuracy on CIFAR10-LT and ImageNet-LT, respectively. The code for this work is available at https://github.com/dgshin21/RNA.