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

2021-08-05 · ICCV 2021 10 · Jiarui Cai, Yizhou Wang, Jenq-Neng Hwang

One-stage long-tailed recognition methods improve the overall performance in a "seesaw" manner, i.e., either sacrifice the head's accuracy for better tail classification or elevate the head's accuracy even higher but ignore the tail. Existing algorithms bypass such trade-off by a multi-stage training process: pre-training on imbalanced set and fine-tuning on balanced set. Though achieving promising performance, not only are they sensitive to the generalizability of the pre-trained model, but also not easily integrated into other computer vision tasks like detection and segmentation, where pre-training of classifiers solely is not applicable. In this paper, we propose a one-stage long-tailed recognition scheme, ally complementary experts (ACE), where the expert is the most knowledgeable specialist in a sub-set that dominates its training, and is complementary to other experts in the less-seen categories without being disturbed by what it has never seen. We design a distribution-adaptive optimizer to adjust the learning pace of each expert to avoid over-fitting. Without special bells and whistles, the vanilla ACE outperforms the current one-stage SOTA method by 3-10% on CIFAR10-LT, CIFAR100-LT, ImageNet-LT and iNaturalist datasets. It is also shown to be the first one to break the "seesaw" trade-off by improving the accuracy of the majority and minority categories simultaneously in only one stage. Code and trained models are at https://github.com/jrcai/ACE.

📄 PDF Abstract BibTeX arXiv:2108.02385

Code (0)

등록된 구현이 없습니다.

Tasks

Long-tail Learning

Three Heads Are Better Than One: Complementary Experts for Long-Tailed Semi-supervised Learning

2023-12-25 · Chengcheng Ma, Ismail Elezi, Jiankang Deng, WeiMing Dong 외

We address the challenging problem of Long-Tailed Semi-Supervised Learning (LTSSL) where labeled data exhibit imbalanced class distribution and unlabeled data follow an unknown distribution. Unlike in balanced SSL, the g…

Nested Collaborative Learning for Long-Tailed Visual Recognition

2022-03-29 · CVPR 2022 1 · Jun Li, Zichang Tan, Jun Wan, Zhen Lei 외

The networks trained on the long-tailed dataset vary remarkably, despite the same training settings, which shows the great uncertainty in long-tailed learning. To alleviate the uncertainty, we propose a Nested Collaborat…

Image Classification Long-tail Learning

LPT++: Efficient Training on Mixture of Long-tailed Experts

2024-09-17 · Bowen Dong, Pan Zhou, WangMeng Zuo

We introduce LPT++, a comprehensive framework for long-tailed classification that combines parameter-efficient fine-tuning (PEFT) with a learnable model ensemble. LPT++ enhances frozen Vision Transformers (ViTs) through …

Mixture-of-Experts parameter-efficient fine-tuning

Self-Supervised Aggregation of Diverse Experts for Test-Agnostic Long-Tailed Recognition

2021-07-20 · Yifan Zhang, Bryan Hooi, Lanqing Hong, Jiashi Feng

Existing long-tailed recognition methods, aiming to train class-balanced models from long-tailed data, generally assume the models would be evaluated on the uniform test class distribution. However, practical test class …

Image Classification Long-tail Learning Test Agnostic Long-Tailed Learning

Balanced Product of Calibrated Experts for Long-Tailed Recognition

2022-06-10 · CVPR 2023 1 · Emanuel Sanchez Aimar, Arvi Jonnarth, Michael Felsberg, Marco Kuhlmann

Many real-world recognition problems are characterized by long-tailed label distributions. These distributions make representation learning highly challenging due to limited generalization over the tail classes. If the t…

Long-tail Learning Long-tail Learning on CIFAR-10-LT (ρ=100)Representation Learning