Decoupled Optimisation for Long-Tailed Visual Recognition

When training on a long-tailed dataset, conventional learning algorithms tend to exhibit a bias towards classes with a larger sample size. Our investigation has revealed that this biased learning tendency originates from the model parameters, which are trained to disproportionately contribute to the classes characterised by their sample size (e.g., many, medium, and few classes). To balance the overall parameter contribution across all classes, we investigate the importance of each model parameter to the learning of different class groups, and propose a multistage parameter Decouple and Optimisation (DO) framework that decouples parameters into different groups with each group learning a specific portion of classes. To optimise the parameter learning, we apply different training objectives with a collaborative optimisation step to learn complementary information about each class group. Extensive experiments on long-tailed datasets, including CIFAR100, Places-LT, ImageNet-LT, and iNaturaList 2018, show that our framework achieves competitive performance compared to the state-of-the-art.

Authors

• Cong Cong School of Computer Science and Engineering, University of New South Wales, Sydney, Australia
• Shiyu Xuan National Key Laboratory for Multimedia Information Processing, School of Computer Science, Peking University, Beijing, China
• Sidong Liu
• Shiliang Zhang State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University
• Maurice Pagnucco School of Computer Science and Engineering, University of New South Wales, Sydney, Australia
• Yang Song

Keywords

• CV: Object Detection & Categorization
• ML: Multi-class/Multi-label Learning & Extreme Classification