Imbalanced_SAM

This is an official implementation of "ImbSAM: A Closer Look at Sharpness-Aware Minimization in Class-Imbalanced Recognition" (Accepted by ICCV2023).

Abstract

Class imbalance is a common challenge in real-world recognition tasks, where the majority of classes have few samples, also known as tail classes. We address this challenge with the perspective of generalization and empirically find that the promising Sharpness-Aware Minimization (SAM) fails to address generalization issues in class-imbalanced recognition. Through investigating this specific type of task, we identify that its generalization bottleneck primarily lies in the severe overfitting for tail classes with limited training data. To overcome this bottleneck, we leverage class priors to restrict the generalization scope of the class-agnostic SAM and propose a class-aware smoothness optimization algorithm named Imbalanced-SAM (ImbSAM). With the guidance of class priors, our ImbSAM specifically improves generalization targeting tail classes. We also verify the efficacy of ImbSAM on two prototypical applications of class-imbalanced recognition: long-tailed classification and semi-supervised anomaly detection, where our ImbSAM demonstrates remarkable performance improvements for tail classes and anomaly.

Enviroment

• PyTorch

Datasets

The train.py will automatically download the CIFAR100 into ./datasets dir.

train

python scripts\train.py \
    --imb_factor ${IMBALANCED FACTOR} \
    --opt ${OPTIMIZER in ['sgd', 'sam', 'imbsam']} \
    --rho ${RHO_IN_SAM OR IMBSAM} \
    --eta ${ETA_IN_IMBSAM} \

Results on CIFAR100-LT (IF=100)

Optimizer	Many	Med.	Few	ALL
SGD	77.8	46.6	13.5	47.6
SAM	79.7	49.3	12.4	48.9
ImbSAM	75.9	53.5	19.4	51.1

The trained model is available at OneDrive, which achieves better performance (ALL Acc of 51.6) than reported results in our paper.

Implementation with Huawei MindSpore

See ./imbsam_mindspore

Thanks to

• ASAM
• WB