Long-tailed Recognition by Routing Diverse Distribution-Aware Experts

Natural data are often long-tail distributed over semantic classes. Existing recognition methods tackle this imbalanced classification by placing more emphasis on the tail data, through class re-balancing/re-weighting or ensembling over different data groups, resulting in increased tail accuracies but reduced head accuracies. We take a dynamic view of the training data and provide a principled model bias and variance analysis as the training data fluctuates: Existing long-tail classifiers invariably increase the model variance and the head-tail model bias gap remains large, due to more and larger confusion with hard negatives for the tail. We propose a new long-tailed classifier called RoutIng Diverse Experts (RIDE). It reduces the model variance with multiple experts, reduces the model bias with a distribution-aware diversity loss, reduces the computational cost with a dynamic expert routing module. RIDE outperforms the state-of-the-art by 5% to 7% on CIFAR100-LT, ImageNet-LT and iNaturalist 2018 benchmarks. It is also a universal framework that is applicable to various backbone networks, long-tailed algorithms, and training mechanisms for consistent performance gains. Our code is available at: https://github.com/frank-xwang/RIDE-LongTailRecognition.

PDF Abstract ICLR 2021 PDF ICLR 2021 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Long-tail Learning CIFAR-100-LT (ρ=100) RIDE Error Rate 52 # 5
Long-tail Learning ImageNet-LT RIDE (ResNeXt-50) Top-1 Accuracy 56.4 # 18
Long-tail Learning ImageNet-LT RIDE (ResNet-50) Top-1 Accuracy 54.9 # 22
Long-tail Learning iNaturalist 2018 RIDE Top-1 Accuracy 72.2% # 13
Image Classification iNaturalist 2018 RIDE (ResNet-50) Top-1 Accuracy 72.2% # 22


No methods listed for this paper. Add relevant methods here